Home / Start Here / Project State

AFFIRMOLOGY - PROJECT STATE (read me first, update me last)

Updated Jul 23, 2026 · Affirmology_PROJECT_STATE.md

AFFIRMOLOGY - PROJECT STATE (read me first, update me last)

STATUS (the 3-line answer to "where are we", keep current)

• POST-ORIGIN CONTEXT-CAPTURE CONCEPT + RESEARCH DOC (2026-07-18, this window; Jeff's ask). New priority project PP-1 (Affirmology_PriorityProjects_v1.md): after the Origin listen, a small results-screen box invites one life-area tap + an optional sentence; we parse/filter it; the NEXT gift (a context+chart reading, then a user-selected-form 3rd audio placeholder) is attuned to what they care about, reverses "too generic" doubt with curiosity, drives the mission-video watch, and builds an "it gets me" feeling over week one. Deliverable: Affirmology_ContextCapture_AttunedReading_Concept_v1.docx (house docstyle, 4 tables, hero image) - full buyer-psychology + competitive + impact research (2 subagents, cited). KEY CRAFT RESOLUTION (grounded in context_engine.py's chart-pure rule): Origin stays pure magic; the box comes AFTER the reveal; the second reading LEADS chart-first (more specificity kills "generic") then turns to their named area (chart leads, context follows). NO chart-pure Mirror in the demo sequence: the #2 piece IS the context+chart reading. The live chart-pure Mirror got good response but a recurring "needs action" note (a B job for slot #2); the context+chart, more-actionable version is the A-job replacement, target soon. The #3 gift = a placeholder delivered in phases: first fixed to the ONE form we can do well, then a user CHOICE (gym push / guided or nighttime meditation / EFT / gene keys journey...) once we can do several well, always testing. Build the #2 piece first. Answers Jeff's 255-vs-1000: ~255 shown, generous hidden cap, one optional field, example placeholder. Impact estimate: activation +25-50%, D7 +20-40%, demo->beta ~+40-60% mid-case. Plumbing mostly EXISTS: the Mirror-drip (shipped today) is the delivery mechanism; the engine already has current_context capture (app-side) to surface on the demo results page. NEW to build: the results.html box + a demo write endpoint + the parse/filter pass + the attuned-reading craft variant + video-watch reveal + the movement/gym audio (needs a new music bed). Cowork owns the craft/copy/QC; Code owns engine/site; DATA owns site/profiles. Cowork handoff paste-block delivered in-chat.
• MIRROR-DRIP LIVE (2026-07-18, this window; per Affirmology_MirrorDrip_Spec_v1.md) - DEPLOYED + ENABLED (studio commits 8e5bb79 + fix a222902 on main, Render redeployed; Jeff: the Mirror is approved and has gone to 20+ people, so every NEW demo-taker gets theirs ~3 days after their Origin, not waiting for Aug 8). A few days after a person's Origin Soul Song delivers, the system auto-renders + delivers their Mirror (the built-in re-engagement beat), no manual batch. Build: new durable MirrorDrip table (api/db_models.py, auto-creates via create_all on boot); new api/mirror_drip.py (schedule_drip written from the Origin-delivered hook in demo_render.py; run_due worker that reuses demo_render._run_batch_mirror(only_emails={person}, limit=1) so EVERY existing batch gate applies - founder/hold/incomplete/duplicate/already-gifted/credit-abort - plus pre-render suppression + _has_mirror gates; per-pass daily cap + spend cap + attempt-based retry/backoff, no silent drops; an in-web-service boot poll worker start_worker(manager) holding the manager since a Mirror render needs the render path; admin/test router: POST /api/mirror-drip/run, GET /status, POST /schedule-test (5-min debug delay for end-to-end verification), POST /backfill). Wired in api/main.py boot block; render.yaml declares the knobs. GATES/CONFIG (plain env, no deploy to change): MIRROR_DRIP_ENABLED (=true, LIVE per Jeff 2026-07-18), MIRROR_DRIP_DELAY_DAYS (3), MIRROR_DRIP_DAILY_MAX (25), MIRROR_DRIP_MAX_COST_USD (5), MIRROR_DRIP_WORKER (worker kill-switch). VERIFIED: py_compile clean; 19-assertion logic test (scratchpad/verify_drip_logic.py); AND live end-to-end on the deployed cloud - /api/mirror-drip/status returns enabled=true, delay 3 days; a schedule-test row for a founder was processed by /run and correctly SKIPPED (sent:0, skipped:1, nothing rendered or emailed), proving the queue write + worker + delegation to _run_batch_mirror + the safety gates all fire. SAFETY GATES on every fire: suppression (unsubscribes) + _has_mirror (never double-send) + the batch's founder/incomplete/duplicate/credit-abort. NOTE: the drip does NOT read the local .mirror_hold file (cloud can't); its live gates are the ones just listed. The stale "HOLD ALL" note inside .mirror_hold (2026-07-03) is OUT OF DATE (Jeff has sent the Mirror to 20+) and should be cleaned up (DATA lane) so it stops misleading future sessions; it does not affect the drip. Follow-up polish (not blocking): a per-person local-morning send window (spec section 3 optional); the est_spent-only-on-real-render fix is already in (commit a222902).
• APP OFFLINE DOWNLOADS HARDENED + MADE PROACTIVE (2026-07-17, this window; Jeff: "make sure the app creates local downloads of ALL audios we created for them, so a user off wifi can still listen"). Finding: a full offline download-and-play system ALREADY EXISTED in the affirmology-app submodule (shared/src/player.ts: downloadAudio/localUri/isDownloaded, and playAudio already prefers the local file over the mp3_url stream) but did NOT reliably meet the goal. THREE real gaps fixed (Jeff chose to keep it INVISIBLE, no UI): (1) INTEGRITY: downloadAsync recorded any status<400 as "downloaded" with no size check, so a truncated/error-page/empty body could masquerade as a good offline copy and play as garbage with no wifi. Hardened: wrap downloadAsync in try/catch (it REJECTS on network drop per SDK 56 legacy docs), require the saved file exists AND size >= 16KB, else delete the partial + return null (unrecorded, retried next run); localUri now also size-gates so any stale partial self-heals. (2) COVERAGE: downloads only fired inside the Listen/Library tab render loops, so nothing saved unless the user opened those tabs. Added syncDownloads(categories?) in player.ts (pulls the whole library, downloads all missing, idempotent, never throws, returns {total,saved,failed,alreadyLocal}) and call it once on app launch from the authed app/(tabs)/_layout.tsx mount, independent of which tab shows first (Library is href:null in the Founding Preview). Both tab loops now call the single canonical syncDownloads(cats) instead of duplicated inline loops. (3) verified: npx tsc --noEmit exit 0, plus a 12-assertion logic simulation (scratchpad/verify_offline_logic.mjs) proving the integrity gates + sync accounting (good mp3 recorded; tiny-200/http-error/network-throw all rejected+cleaned+unrecorded; stale partial self-heals; sync buckets sum to total; idempotent re-run saves 0). SHIPPED 2026-07-18 (Jeff: "commit + kick EAS build now"): committed shared submodule a5309de + app repo 09ff31a (submodule bump), then EAS production build 37 (Build ID 6d3e809c) built + auto-submitted to TestFlight (App Store Connect 6784908433); Apple processing at write time, Jeff taps Update once done. NOTE: the shared canonical remote (affirmology-shared) push was rejected (denyCurrentBranch, non-bare + main checked out) so a5309de lives locally only, which is all EAS needs; reconcile the canonical repo later. Still to confirm: true native offline playback on-device (the Xcode/TestFlight run, per app AGENTS.md). No new native deps added; no cellular/wifi gate (downloads over any connection, matching the guarantee-availability goal; a wifi-only toggle is a later option needing netinfo).
• EMAIL OPT-OUT / UNSUBSCRIBE SYSTEM BUILT (2026-07-14, DATA chat). Kara Baruzzini asked off the email list -> DONE (engine opt-out affirmology-studio/email_unsubscribed.txt + .beta_invite_hold UNSUBSCRIBED section + .mirror_hold PERMANENT marker + CRM docs annotated). NEW systematic unsubscribe at the single send choke point: api/suppression.py (committed opt-out file + runtime disk JSON + HMAC-signed tokens + on-brand footer), api/emailer.py send_email() now drops suppressed recipients (to+bcc) and auto-appends a working unsubscribe footer to every single-recipient customer HTML email (owners excluded), and public GET /api/unsubscribe (main.py) records opt-outs. Verified locally + py_compile clean. DEPLOYED 2026-07-16 (CODE chat): studio commit 1553553 (only the 4 unsubscribe files; the tangled PDF-report v2 WIP + a stray 19MB corpus_baseline.db were deliberately left uncommitted) pushed to origin/main -> Render auto-deploy. STILL OPEN: the footer link base defaults to demo.affirmology.ai but the demo worker does not route /api/unsubscribe (404), so set Render env UNSUBSCRIBE_BASE_URL=https://affirmology-studio-api.onrender.com (no wrangler) OR add a worker route. Suppression FILTERING is live immediately; only the clickable footer link waits on that env var. See _CHAT_SYNC 2026-07-14 + 2026-07-16 CODE blocks. ALSO 2026-07-16 (CODE): affirmology-site now auto-deploys via GitHub Actions (.github/workflows/deploy.yml, commit 95ca179) to end the manual npx wrangler deploy step; needs 2 one-time repo secrets (CLOUDFLARE_API_TOKEN + CLOUDFLARE_ACCOUNT_ID).
• BUILD BRIEF RECONCILIATION (T9-T14 v2) COMPLETE (2026-07-13, DATA chat). Jeff's full brief arrived post-ship; the build now matches it exactly: app_config.py single-source tunables (env-overridable), Intention/Block terminology everywhere (context = umbrella only), regular 1.5x tier multipliers + generation unlocks, writeBirthTime named tool, Aries v2 (aliased categories, switchable rules, rolling-week grace, month archives + lifetime record, reward-table-driven diverse rewards with spirit-guide rung 1 + change gate), Intention/Block free unlimited changes + apply-now at 10 credits + Session Setup per audio, and the pace PIVOT to real-time playback time-stretch (0.8-1.2, no re-render, no pitch shift, native-by-family + personal dial). Unit-verified server-side; app rides build 36. Open items left switchable: dayCountsRule, constellation bases, reward table contents.
• PART B (T9-T14) COMPLETE, BUILD 35 TO TESTFLIGHT (2026-07-13, DATA chat). Credit ledger live (three buckets, spend order, tier multipliers, SamCart seam unbuilt); birth-time write-back closes the build-34 gap (pinned time -> profile + chart rebuild + queued fresh Origin); Aries engine (monthly fills, one-grace streak, idempotent rung rewards into earned credits + gift-unlock events); context engine (chart-pure families protected, first-free-then-paid changes, context-tagged audios, revisit surface); Home tab assembled (moon/eclipse ephemeris feed, inspiration, journaling, context card); voice pacing plumbing (structure-family defaults, personal dial at both synthesis paths, character-only votes, retirement swap). All unit-verified; details in _CHAT_SYNC 2026-07-13 Part B block.
• T1-T8 BUILD BATCH COMPLETE, BUILD 34 TO TESTFLIGHT (2026-07-13, DATA chat, Jeff's overnight batch). The Wheel rebuilt manifest-driven per the FINAL alignment (Sun home with favorites inside + state dial, ring named by REPRESENTS, star pattern only on scroll-up, interface tray, nameless stars identified by their audio, x/7 listen counters on a new play_completed event spine); You hub (nickname/pronunciation, Current Context feeding generation via Profile.current_context, highlights, Birth-Time card); People screen (chart/described/linked-by-consent share codes + inbox + revocation + photo); reusable SurveyDeck folding answers into the Understanding via distill; oracle-handoff TrackSheet (Continue with Sophia carrying track_brief, resume via survey-state) live on the Wealth Claim; feedback instruments (2-tap landed at audio end + celebration + voice reply + block-intensity 1-10 intake/day-7); birth-time rectification chat entry (write-back of the pinned time still an ops step). All simulator-verified; studio endpoints deployed; details in _CHAT_SYNC 2026-07-13.
• WHEEL BRIGHTNESS + FEATHERED PLATES LIVE (2026-07-13 late, DATA chat, per Jeff's two art calls + the Prototype_v4 reference). All 12 figures republished as v3_ R2 keys with a luminance lift + an elliptical alpha feather to zero at the plate edge (kills the rotated-plate paper edge on Pisces/Virgo); alignment.json repointed (FINAL-3-bright-feathered), hot-swapped to devices with NO build, CDN-verified. Brightened sky at wheel/v2_blank_sky.png + app code (8463611, rides build 34) lifts the wheel base to the prototype emerald and reads the sky key from alignment.json. Build 33 (Sophia freeze fix + Hermes shifter + chat pulse) is at Apple processing.
• SOPHIA FREEZE ROOT-CAUSED (SIM-VERIFIED) + ORACLE CHAT GROUNDED IN DEEP UNDERSTANDINGS (2026-07-12 late night, DATA chat). (1) Sophia's animation froze on device because state transitions loaded video sources synchronously over the network; rewritten to four preloaded players, transitions are play+crossfade only; the blaze now pulses on her newest chat-response avatar (Jeff's call); verified by recording the iOS simulator and inspecting frames; app commit 829175f, riding build 33 (fired to TestFlight with the Hermes shifter). (2) Jeff's Chiron test exposed the oracle chat pulling zero chart facts and a first-2500-char Understanding slice; now grounds in the canonical stored Profile chart + prefers profile.deep_understanding with a core-first 8k slice (studio cc2d8c0+bb59965+6c1f944+28efd68, deployed). Deep six-lens docs now live per-person: Jeff + Sol migrated, Colin built cloud-side (streamed, ~$0.30, 54k chars). Audit: all 38 stored Understandings intact, nothing was ever truncated in storage.
• BUILD-32 TEST SESSION FIXES + HERMES SHIFTER (2026-07-12 late, DATA chat, Jeff testing on device). (1) Wheel figures looked OLD in the app: the CDN was serving stale bytes on the overwritten keys; clean art republished under wheel/figures/v2_ keys + alignment.json repointed, verified clean through the CDN, fixed with NO build (new rule in _SHARED_CONTEXT: never overwrite an art key). (2) Hermes/Thoth shape-shifter chat avatar built per the OracleAvatars handoff (app commit 1f78cbe, rides the next build; pre-reversed mp4 made locally). (3) Jeff's feedback logged: the Wheel needs a real rotating-wheel redesign ("bigger image map", swipe-down-to-Sun missing) and better nav icons (Jeff sourcing examples for Cowork). (4) Jeff's call: NO daily email until Aug 13; the Wave 1 email ask is on hold.
• BUILD 32 TO TESTFLIGHT + NEW INTAKE FORM LIVE (2026-07-12 night, DATA/engine chat). App build 32 (the Wheel at Jeff's FINAL alignment + 12 clean figures + all 5 pronunciation screens, shared @ 22e1004) built on EAS and auto-submitted to TestFlight (Apple processing; Jeff just taps Update). The new intake form (name_called, birth_time_known fix, 3-tier name validation posting name_flag) is DEPLOYED on demo.affirmology.ai, and a silent field-dropper in the demo worker was found + fixed (its /api/render job object hand-picked fields; now forwards name_called/name_birth/birth_time_known/name_flag to /render-sacred-audio; worker 38d762d2, site 2a2bcfd0, verified live). Jeff's canon chant confirmed = take 4, canonical R2 file byte-verified. Still open: Cowork owes the Wave 1 daily email copy (Affirmology_CoworkHandoff_NameFlags_Wave1Email_v1.md); site form deploy no longer blocks anything.
• APP EXPERIENCE SUPERCHARGE SESSION COMPLETE (2026-07-11 into 7/12 overnight, Cowork app-experience chat with Jeff live). Affirmology_AppExperience_MasterPlan_v1.md is FINAL-for-now through section 16 and is the lead reference; highlights decided tonight: (1) THE WHEEL locked from Jeff's pen sketch (saved in WheelConcepts): implied wheel never drawn, sun at center (the Sun IS the person, holds favorites = the TREASURE BANK; earned audios stay accessible after membership ends), 12 zodiac wedges scrolled horizontally with the focused constellation upright, swipe down = the Sun, swipe up = beyond-the-rim journeys; star spec = empty stars are thin gold OUTLINES with inner shimmer (variant 3), vessel-fill to equal full shine, LINES follow aliveness (extend on first fill, brighten with fill); figures go silhouette > gold-mist > formed starlight (approved Aries progression sheet). (2) ZODIAC THEME MAP drafted with real star counts: Aries=Initiation (4 stars, live), Taurus=WEALTH, Gemini=Toolkit, Cancer=Family, Leo=Confidence, Virgo=Health/Craft, Libra=Relationships v1, Scorpio=DEPTH PSYCHOLOGY, Sagittarius=GENE KEYS Golden Path (main wedge), Capricorn=Career, Aquarius=Human Design, Pisces=long journeys+dreams (name TBD); outer tier = Chiron healing, synastry constellations, challenges (breakable, opt-in, 7-straight/14/month ladder). (3) THE CHANT LANE IS LIVE: Eleven Music API (new full-access key, Jeff has it, Render env swap pending with Code) rendered Gregorian chants of Jeff's name for ~pennies; Jeff loved v1; verdicts: name must ALWAYS be audible (render a cappella + mix OUR bed at ~0.18, never let the model add instruments), generation is a lottery so render 3-5 takes and keep ONE by ear; NOT an Origin prefix (tested, register clash); LOCKED FLOW: the chant is the FIRST-OPEN WELCOME (plays over the wheel flash-forward, behind a SOUND GATE: begin button requires volume up, playsInSilentModeIOS beats the mute switch, ends with one tap 'Did you hear your name?' = the per-person QC gate since STT cannot certify sung melisma) and RETURNS at the Aries completion ceremony where the chosen guide GIFTS it into their Sun to keep. (4) Other locks: Event Audio killer feature (one custom audio ON the day of each new/full moon + big events, anticipated unlocks; Night Tide refreshes biweekly on the lunar rhythm); guide cadence = choose at ceremony then a new guide ~monthly (Sophia stays the constant synthesis layer; guide = the buddy); no separate native Atlas app (role-gated CREATE tab wearing the ATLAS name + web/PWA lab); Explore = icon chip rail + panels, The Sky default with idle drift-home; readings = podcast shelf; Get Audios chip killed (requests live in context pages); beta feedback sunsets Aug 13 (customer launch); TIMELINE: Mirror email Sun AM, join text Sun PM, deeper unlock Tuesday. (5) OVERNIGHT: Code has the 10-item build order (Explore scaffold, wheel component with R2 art manifest contract, welcome flow, Listen reshape with registry names, favorites, readings shelf, feedback boxes, status banner, context pick); the full 12-constellation dormant-sky art set + the Sun rendered via Nano Banana into Higgsfield for morning curation (Taurus re-run facing LEFT per the real star mapping). Explore-tab interactive mockup v2 in WheelConcepts.
• BETA ACTION MAP v2 WRITTEN + CORPUS OVERWRITE INCIDENT DIAGNOSED (2026-07-11, Cowork briefing chat, docs only). (1) Affirmology_AudioPossibilities_Map_v2.md supersedes v1 as the WORKING map: the ranked Wave 1/2/3 build order for beta (Wave 1 = Daily Pair S5+S11 as locked structures, S2, S1 template, S9 day-6 wow, minimal journey orchestration via library ingest + daily email, and the feedback instruments; floor = Daily Pair + feedback taps), two NEW structures from the 7/11 corpus briefing (S20 The Belief Surgery: affect bridge > reconsolidation pairing > Ericksonian identity seal, week 2+ gated; S21 The Identity Gym: Maltz rehearsal + McColl micro-action + evening vote acknowledgment), two new template rules (every structure DECLARES its induction, the 0.95 corpus entry; reconsolidation order is law: reactivate > contradict felt > repeat), the HABIT ENGINE (the Update Install frame: visible SOS version number, every audio = a patch with build-report-generated patch notes, play = install, week one = the v1.0 to v2.0 upgrade, "1 update pending" as the native Zeigarnik loop; habit stacking onboarding, identity-named streaks with one grace, anchor-word carry system, locked ethics line), the ASSIMILATION ENGINE (the closed 7-step loop mine > offer > name > measure > update > synthesize > per-person; only steps 4 rating-capture and 5 feedback-to-craft-confidence write-back are unbuilt, both small), the testing kit, and NOW/NEXT/LATER. v1 stays as the palette reference; TechniquesLibrary_v1 untouched (engine-ingested). (2) CORPUS INCIDENT: the 7/11 nightly cron researched a 42-entry somatic/attachment/IFS-deep batch and pushed the enriched corpus.db ~09:05 UTC; a manual corpus.db upload at 09:36 UTC OVERWROTE it (R2 LastModified evidence). Entries not lost as work items: all 42 sit pending in the backlog (56 pending total) and re-mine at the nightly batch rate, or one manual cloud research_lane run catches up. NEW RULE (also in the map, Section 1): never manually upload corpus/corpus.db to R2 without pulling current + checking LastModified against the last cron push (~09:00 UTC); merge, never replace.
• SOPHIA TRANSFORMATION MECHANIC SHIPPED IN BUILD 26 + FOUNDING TEXT LINKS COMPLETED (2026-07-11, CODE). (1) SOPHIA MECHANIC: the oracle-chat guide's activation is now a VISIBLE whole-avatar TRANSFORMATION (resting -> wild starlight hair when she starts speaking, wild -> resting when she finishes), replacing the too-subtle eye-glow. Rewrote affirmology-app/src/SophiaHero.tsx as a two-layer ~250ms-crossfade 4-clip state machine (RESTING loop / ACTIVATE once / ACTIVE wavy loop / RETURN once), driven by the speaking (busy) prop, using expo-video playToEnd + a safety timeout for the one-shots. Clips bundle for instant play AND load from R2 (media.affirmology.ai/avatars/manifest.json v2 -> sophia_rest/activate/active/return.mp4, uploaded from Product/SophiaHandoff/animations/) so her look is SWAPPABLE WITH NO APP BUILD. Typecheck clean. Commits: app 5645995, shared d8aa4bc (adds the neutral team/Affirmology name for the crisis/legal icon). Also carries the neutral team icon (OracleBeacon) + Hestia name. (2) BUILD 26: EAS prod build --auto-submit -> App Store Connect app 6784908433, build 26 (id 2b82eb1b) VALID, added to the external "Affirmology Beta" group (public link https://testflight.apple.com/join/nDJacr85), and its beta review is already APPROVED (auto-approved as a subsequent build on an already-approved external group). Jeff (internal tester jalanparker@gmail.com, the sole member of the internal Team(Expo) group, allBuilds=True) gets it automatically; the founding group can install NOW via the public link. Sol is NOT yet an internal tester (gets it via the founding group). Reusable ASC helper: scratchpad/asc.py (status/promote/verify26b/testers) via scratchpad/ascvenv (PyJWT/cryptography/requests; key YGGKMDU9H2 / issuer 19bd7795 / p8 at affirmology-app/asc-api-key.p8). (3) FOUNDING TEXT LINKS COMPLETE: reconciled the FULL studio roster (not the old invite log) into the demo worker PEOPLE map (affirmology-demo-site/worker/src/index.js, 31 codes, deployed) so every founding person has a tracked /r/<code>?to=commit link (302 -> agreement 200 -> /api/beta-signup, records commit against the token->email). Fixed the /agreement 404 (re-deployed affirmology-site). AUDITED all 31 tokens resolve to the right email; FIXED two that were wrong via a NEW admin endpoint POST /api/admin/fix-demojob-email (studio commit 5c36540, deployed): josh resolved to jeff@jeffparker.love (would credit Jeff) and jeanette to jnetrios@gmail.con (typo) -> corrected + re-verified. Final 29-person text-now list + the do-not-recruit list (Hague, Hilsbos, Cleveland, Pruitt; Amy Dholakia pending Sol) in Affirmology_TextableLinks_PerPerson_v2.md. Randy Singer link works but his birth data is weak (fix before rendering). OPEN: Colin still needs a Mirror rendered (+ confirm his Origin Soul Song exists); the chatbot's context-GROWTH write-back (distill conversation -> update learned_info/context_store) is still unbuilt (backend, no app rebuild needed).
• MISSION VIDEO UPDATED WITH THE SOL UPDATE (2026-07-09, DATA). Jeff dropped Affirmology Sol Update.mp4 (618MB, 1080p30, 6:51 - the mission/investor video with the updated Sol segment) in the folder and asked to update the live mission video with it. Web-encoded it the proven web-light way (ffmpeg libx264 preset slow crf22, maxrate 2000k, AAC 128k, +faststart) -> Affirmology_Sol_Update_web.mp4 105MB (83% smaller, full 411.6s, audio intact mean -26.8dB), then OVERWROTE the R2 key experience/mission-video.mp4 (bucket affirmology-media, creds from affirmology-agent/.env). Verified LIVE: https://media.affirmology.ai/experience/mission-video.mp4 now serves 105,316,211 bytes, last-modified today, HTTP 206 byte-range streaming OK, cf-cache DYNAMIC (no stale cache, no purge needed). This is the same key already wired into result.html's "Where this is going" section, so the live results page picks it up with no code/site change. (Prior web encode was Mission_Video_web.mp4 88MB, now superseded.)
• ELEVENLABS GRANT READY TO SUBMIT (2026-07-09, Cowork). Jeff decided: submit with the demo link + reviewer walkthrough, no new affirmology.ai page. A reviewer NDA-bypass was added to affirmology-site/index.html (https://demo.affirmology.ai/?review=one-sky-2026 skips the gate; used ONLY in the grant application, remove after the decision). Submit-ready answers in Affirmology_ElevenLabsGrant_Application_v2.md, including the note that the demo intro film narration is pre-produced while every personalized Soul Song render is ElevenLabs. AWAITING: Jeff pushes the index.html commit (bypass is local-only until pushed), verifies the link in a private window, runs one clean demo render, submits at elevenlabs.io/startup-grants. Queue item S6 updated to in-progress.
• AUDIO POSSIBILITIES MAP v1 WRITTEN (2026-07-09, Cowork briefing chat, docs only, no engine code). Jeff asked for the full possibility space of what the mined craft corpus (339 craft entries, 213 practitioners) can become for beta testing and journey sequencing. Delivered Affirmology_AudioPossibilities_Map_v1.md: the nine building-block families, the universal vs context-dependent vs chart-dependent dependency sort, the product containers, 19 named structure ideas indexed S1-S19 (Rewire Arc, Meet the Protector, Night Install, Inner Council Session, Morning Claim, Threshold Reading, The First Week journey, etc.), a proposed day-by-day First Week beta journey with two A/B alternates, sequencing rules, the testing/feedback frame, and next research batches (backlog is at 0 pending, needs feeding). Doc is written as a HANDOFF for another Claude process to sort/prioritize: it opens with a "FOR THE SORTING CLAUDE" section (operating rules + machine-sortable index with status and metric per structure). Nothing built yet; sorting and build order are the next step.
• SOL'S CLAUDE DESKTOP ONBOARDING + JPL-KNOWLEDGE PACK STAGED (2026-07-09, Cowork). Sol is on Claude Desktop (chat only, older Mac, Pro plan). Her Claude tried to clone github.com/jeffparkerlove22/jpl-knowledge and hit an auth wall: the repo was PLANNED (StudioDeploy runbook + TeamAI onboarding both reference it, link "to come") but never created. Cowork staged the full repo-ready pack at jpl-knowledge/ in this folder (README = Affirmology_KnowledgePack_v2.md, refreshed to July 2026 incl. the corpus-count reconciliation 12,900 June -> ~25,500 now; craft/, plan/, system/, sol/ incl. her chart facts + wealth transcript + Suno prompts, from-sol/ outbox, .gitignore safety net). Runbook equity-grep ran CLEAN (only benign hits: "harvesting", and the Atlas doc SAYING collaborators cannot see equity). BusinessPlan_v5 deliberately EXCLUDED from the repo (ownership content; Colin gets this repo later); share it with Sol by email. Full guide with Jeff's create/push/invite steps + Sol's Project/connector setup: Affirmology_SolClaudeDesktop_Setup_v1.md. OUTCOME (same day): the repo DID already exist (first pass: Jeff/Sol/Colin profiles/, investor-opportunity/ with ExecSummary v3 + InvestorBrief incl. round structure and valuation per Jeff's inner-circle stance, agent designs, craft docs). Cowork merged the v2 pack into it (rebase conflicts on README + .gitignore resolved; README v2 now reflects the FULL repo incl. profiles/ + investor-opportunity/; runbook .gitignore safety net merged in). PUSHED LIVE by Jeff (commit 6fb0a12 on main). REMAINING (Jeff): invite solballard@gmail.com as collaborator with Write access, send Sol Part B of the guide.
• CLOUD UNDERSTANDING-BUILD ENDPOINT LIVE (2026-07-08, DATA) - the priority infra fix from the 2026-07-06 standing rule (ALL Understanding generation must run cloud-side, never Jeff's laptop) is DONE and verified live on Render. New module affirmology-studio/api/understanding_build.py + three routes on the demo/profile router (commits b4b6c0e + edc8703, both deployed): (1) POST /api/build-understanding {person, force?, wait?} fire-and-forget (daemon thread) build for one person: reads their STORED Profile (no chart recompute, honors the ProfileCanonicalSource principle), runs ONE grounded Sonnet-4.6 pass over the locked 12-field schema (salient_findings, convergences, aha_moves, wound_gift_gap, mechanisms, low/high road, metaphor_bank, one_liners, fear_preemptions, timing, context_store, learned_info), writes profile.understanding via set_understanding; (2) POST /api/build-understanding-batch {confirm?, only?, skip_existing?, limit?, pace_seconds?, force?} DRY-RUN BY DEFAULT (returns count + est_cost_usd + candidates), confirm:true runs it in the background one-at-a-time with a gentle pace, skip-existing, and NO blind retries (cost discipline enforced in code); (3) GET /api/build-understanding/status live progress. Auth = X-Affirmology-Auth: <DEMO_AUTH_TOKEN> (same as the sibling /api/profile + /api/profile-understanding). The render/safeguards chat's Understanding generator can now retire any laptop-bound script and call this instead; the existing write-path is unchanged, this adds the server-side GENERATE step. VERIFIED not just at HTTP-200: the first real build (Josh Parini) caught + fixed two defects (doc truncated at section 8 of 12 because max_tokens 8k was too small, now 16k via env UNDERSTANDING_MAX_TOKENS; and it emitted em dashes despite the instruction, now deterministically stripped to enforce the hard brand rule). Josh rebuilt clean: 12/12 sections, 0 em dashes, 7,008 words, 118 grounded chart markers, pattern-first-then-proof craft. COVERAGE: 35/35 profiles now have an Understanding (no_understanding=0). The 4 that were missing (Kara Baruzzini, Josh Parini, Elmira Shahzeidi, Rodrigo Gayoso) were all built via this endpoint 2026-07-08 (Jeff greenlit the batch), each verified clean: 12/12 sections, 0 em dashes, 7.6-8.3k words, 85-158 grounded chart markers. Total spend ~$0.80. SIDE FINDING: the Colin + Jeanette duplicate merges flagged in _CHAT_SYNC (2026-07-06) were ALREADY DONE (live admin overview: duplicate_people=0, zero orphans, both keepers healthy with chart + understanding); nothing to merge.
• CRAFT KNOWLEDGE BASE WIRED INTO SCRIPT GENERATION (2026-07-05, DATA). script_generator.generate_script now retrieves a live OFFERED craft menu at generation time (_offered_craft_menu, keyed on the audio concept + purpose, n=12), injected as optional craft the writer may weave in where it serves this chart and this structure, never mandatory content. Unlike the per-stage menu it spans every craft area, so the 27 sourced practitioner entries from the overnight batch (open loops and curiosity gaps, identity-based persuasion, pause craft, prosody and mood arcs, Ericksonian patterns) now reach the writer. Gated to the flex/Studio path so the locked Origin/demo container stays byte-stable; no chart-math or structure changes. Root fix: local corpus/corpus.db was STALE (0 craft rows), so the engine saw no craft; re-pulled the current DB from R2 (339 craft rows, 213 area=practitioners) so local renders match the cloud. Verified: search_craft surfaces the practitioner entries (Zeigarnik/open-loop, Pause craft, Prosody arcs, Ericksonian, Identity-based persuasion rank #1 for their queries); pytest 68/68 green; a Studio test render's build report named practitioner techniques including "Identity-based persuasion (selling who they become)", "Joseph Campbell (hero's journey)", "Veritasium and Kurzgesagt narrative hooks", "Self-relevance and prediction error". TechniquesLibrary to-do ticked. NOTE: the code change is in the studio-vendored engine working tree (affirmology-studio/affirmology-agent/.../agents/script_generator.py), tested, ready for the next Render deploy; not yet committed (the repo holds another chat's WIP, so coordinate the commit).
• BACKEND ADMIN CONSOLE LIVE + UNDERSTANDING STORAGE FOR EVERYONE + MARC FIXED (2026-07-05, DATA). (1) New purely-operational admin dashboard at https://studio.affirmology.ai/admin (Cloudflare Access login, then paste the Studio token once) or raw https://affirmology-studio-api.onrender.com/admin. Not a second Atlas: api/admin_site.py + api/admin.html, all gated by the one Studio token. Shows every Profile with health flags (chart / understanding / email / birth-complete), duplicate-people detection, counts; one-click repair: rebuild chart, set email/atlas/understanding, delete, MERGE duplicates into one keeper. Live overview at push: 35 profiles, 8 no-chart, 1 emailless, 2 dup groups (josh parini, colin jacobs). (2) THE UNDERSTANDING now stores on EVERY profile, no founder distinction: new Profile.understanding column + get_profile/GET /api/profile return it + write-path POST /api/profile-understanding {person, understanding} (resolves id|email|atlas handle|name) + profile_report.set_understanding. The render/safeguards chat (Understanding-generator lead) must point its generator WRITE + audio consumer READ at these (handoff in _CHAT_SYNC). (3) Marc Kessler root cause: profiles key by email; his real-email row (mak2001@bellsouth.net) never had a chart + two emailless orphan dupes existed. Fixed build_and_store_profile so an emailless build reuses the same-name row instead of orphaning; consolidated Marc to one row with a full chart (Libra/Virgo/Sag). Commits da7a66b (storage + dedup) + e1a176f (console), both live on Render. (4) Sol connected to both apps: magic-link login sent + on-brand instructions email (Maya + Atlas TestFlight links). Open (Cowork): backfill Sol's 7 library audios (needs her app-account id).
• THE UNDERSTANDING: DECIDED + SCHEMA LOCKED (2026-07-04, Jeff approved Code's recommendation). The per-person deep read stops being a single prose blob and becomes a structured, growing UNDERSTANDING, built INSIDE the Mirror/reading-craft rebuild (not separately). Council builds it once per committed person; every audio/chat after is a cheap shaping (select facets + apply craft); learn-big-things = APPEND, never recompute. Two tiers: Tier 1 light (everyone gets Origin), Tier 2 deep council Understanding on commitment (rec trigger: the "Yes, I'm in" beta reply). Schema v1 (12 fields: salient_findings ranked, convergences, aha_moves, wound_gift_gap, mechanisms, roads, metaphor_bank, one_liners, fear_preemptions, timing, context_store, learned_info) is in Affirmology_FlagshipReading_Proposal_v1.md section 3. One-pager decision 3 marked DECIDED.
• LAUNCH WEEK CONSOLIDATED TO ONE PAGE + BETA EMAIL DRAFTED (2026-07-04, Cowork). Jeff asked for help organizing the sprawl: Affirmology_LaunchWeek_OnePage_v1.md is now THE map (plan, architecture, craft, Jeff's 5 open decisions, owner split, doc map). Key unification: Code's "structured Understanding" = Cowork's "Deep Map"; one name going forward, THE UNDERSTANDING, built inside the reading-craft rebuild (Code + Cowork agree). Craft-menu trim approved (trim + dedupe across stages). Beta pivot per Affirmology_BetaLaunch_RevisedPlan_v1.md: NOT opening 7/4; email today/tomorrow, gift drip through the week, beta opens Fri 7/10. Beta invite email DRAFTED: Affirmology_BetaInviteEmail_v2.html (craft-first framing + one small trip line + human-answers service line, per Jeff; reply "Yes, I'm in" mailto hello@affirmology.ai as the affirmational signature). SEND DEPENDENCY: hello@affirmology.ai must actually receive replies (Cloudflare Email Routing forward) and the send should use a reply-able from/reply-to, NOT the no-reply reports@ pattern. Jeff's 5 open decisions on the one-pager: send timing, tester list, when to build the Understanding (rec: now, with the Mirror rebuild), Tier 2 trigger (rec: the "Yes, I'm in" reply), starting audios set. Cowork still owes: beta agreement text, Sol TestFlight data fix, wiki/FAQ.
• BETA ROLLOUT PLAN SET (2026-07-04, Cowork, Jeff's decisions). Plan: Affirmology_BetaRollout_Plan_v1.md. Beta does NOT start July 4; honor the date with a real email today, stagger personal gift audios across the week, open the beta FRIDAY 7/10 (Jeff back in Miami office late Tue/Wed). LOCKED: (a) beta start Fri 7/10; (b) TODAY 7/4 send the expectations + affirmational-agreement email, aspirational "care and craft" framing, NO trip mention (Affirmology_BetaExpectationsEmail_v1.html, subject "The Founding Constellation is forming", both-ways expectations + one-tap "I'm in" mailto affirmation; test sent to Jeff); (c) The Mirror goes to the FULL list Sunday AM 7/5 after Jeff+Sol QC Fri night (Staci maybe early, probably not). Cadence: Mirror Sun 7/5, a new audio Mon/Tue, the report Wed/Thu, all to the list. CODE items: sync Sol's TestFlight build (currently NO data), load starting audios+beds, build in-app daily-listen + feedback capture + an ML testing loop (ask testers questions, capture prefs, generate variations, engine learns/adapts), generate Mirror MP3s (mirror_url), and remove the leftover Origin Soul Song download button in the next deploy. COWORK items: the gift emails, tester WIKI + short videos, feedback forms/question set, and CO-BUILD the beta agreement wording (first-draft terms in the plan). Jeff steers within ~2 days: starting audios + music beds. DEPLOY NOTE (learned 7/2): the site is a Cloudflare WORKER affirmology-site (Workers Builds) in account e9740eb6, NOT Pages; GitHub auto-build got stuck, so deploys now go out via npx wrangler deploy from the affirmology-site repo (accept detected settings: name affirmology-site, framework Static, output dir .). Results page is live + correct (green Mirror player, welcome video, cover, reorder, scroll-to-top).
• FIRST-READING STRUCTURE + ORIGIN B VARIANT SPEC'D, CRAFT CORPUS ASSIMILATED (2026-07-04, Cowork). The overnight corpus tool delivered 44 storytelling/hooks/persuasion/delivery craft records (docs: Affirmology_StorytellingHooks_Synthesis_ForCode_v1.md + _CraftBatch_v1.md; Code is wiring per-stage craft retrieval into script_generator.py per that doc's Part 4). Cowork fused those with the wealth-transcript techniques + aha registry into: (1) Affirmology_FirstReading_Structure_v1.md, the definitive structure replacing The Mirror's component walk (keep public name THE MIRROR, internal id mirror_recognition_v1): Promise -> Being Seen -> Three Doors (Big 3 convergences, each dip-and-rise with mechanism, contrast, portable law, practical move, twice-spoken affirmation) -> Install (master-loop payoff, Ericksonian) -> Seal; ONE master symbol threaded (reconciles the corpus one-symbol rule with the metaphor bank), one loop always open, rehook every seam, one crescendo in Door Three, 12-rule QC gate, test-on-Sol-first plan. (2) Affirmology_OriginSoulSong_Bvariant_Spec_v1.md, the origin_soul_song_b_craft A/B: container unchanged, six line-level changes (early convergence thesis line, threaded master symbol, two mechanism lines replacing the most generic lines, one contrast beat, one Ericksonian install line, planted micro-loop paid by the existing close) + the "becoming her" pronoun fix in both arms; deliberately a separate test axis from The Recognition A/B, run as separate arms.
• WEALTH READING TRANSCRIPTS RECOVERED + FLAGSHIP READING / DEEP MAP PROPOSAL (2026-07-03, Cowork, same session as the reading-craft reset below). The two 72-min Master Wealth Reading audios existed ONLY as MP3s on R2 (media.affirmology.ai/library/{jeff,sol}_WealthReading_full_clean.mp3, emailed 6/24 + 6/26); no scripts were stored anywhere. Cowork downloaded + Whisper-transcribed both (~10.3k words each): Affirmology_WealthReading_Transcript_Jeff_v1.md + _Sol_v1.md, now the canonical style exemplars (style shared, content never). Transcript mining found 15 further named techniques (mechanism move, portable law, convergence refrain "same address", fear preemption, somatic checkpoints, jargon demotion, gears timing structure, etc.) AND two render flaws: composer meta-seams spoken aloud ("I have everything I need from Sophia, let me write section two") and Jeff's MP3 truncated mid-sentence at 71:37 -> new QC checks needed. FULL PROPOSAL: Affirmology_FlagshipReading_Proposal_v1.md = (a) Mirror v4 flagship 10-12 min reading, minute-mapped (Threshold -> three Doors -> Seal; convergence spine + technique stack); (b) the DEEP MAP architecture: one Deep Run of focused 1-2-oracle council calls per person -> structured Blueprint Deep Map (Big 3 convergences, salience table, mechanism library, low/high roads, bespoke metaphor bank, one-liner bank, fear preemptions, timing calendar, life context) stored as pinned KB; ALL audios compose from the Deep Map, NEVER raw chart JSON (the profile is a map/router, not a source); (c) tiering: demo = no deep run (Enduring identity + one light salience pass), committed beta/subscribers = full Deep Run overnight on commitment, new topics = cached mini runs per lens; no paid audio ever composes from the basic profile alone; (d) Origin Soul Song gold-script audit (tours placements instead of one convergence thesis, under-used stone/lake metaphor, a few generic lines, no contrast beat, and a PRONOUN BUG: Jeff's approved script ends "I am already becoming her"); apply fixes to The Recognition A/B first.
• READING CRAFT RESET: MIRROR v3 FAILED THE LISTEN TEST, HERO'S STORY IS NOW THE PRODUCT-WIDE STANDARD (2026-07-03, Cowork, from Jeff+Sol's 2026-07-02 car deep-listen). Sol's Mirror v3 was a component-by-component chart walk (sun/rising/moon/HD type/profile, 4 generic sentences each): no ahas, no synthesis, no payoff, too fast, naked jargon, monotone. It ignored Affirmology_AhHaMoves_Registry_v1.md. Meanwhile the two long (~72 min) WEALTH reading audios are the craft bar (epic book opening, sky-on-the-day-born, pattern-first then placement-as-proof, "here's what that means" stepdowns, multi-domain examples, fresh visual symbols, low-road/high-road contrast, salience callouts, pause/breath beats, affirmation-grade one-liners like "not wired for a salary, wired for a movement", passion crescendos, hooks + rehooks/open loops). FULL SYNTHESIS + SOLUTIONS + NEXT-READING SPEC: Affirmology_ReadingCraft_HeroStory_Synthesis_v1.md (unifying theme = the customer's hero story: seen, epic, meaning, path; the 5-step segment template; hooks/rehooks toolkit; Mirror v4 spec = 10-12 min BIG 3 salience reading, depth over comprehensiveness, leave them wanting more; Code action list incl. extending .mirror_hold to everyone, script-level craft QC gate, player speed toggle). NIGHTLY CORPUS RESEARCH BRIEF (paste-ready, storytelling/hooks/rehooks/persuasion/hypnotic delivery, 5 domains + record format): Affirmology_CorpusResearch_StorytellingHooks_Brief_v1.md. MISSING ASSET (Code): the two long wealth audio TRANSCRIPTS are NOT in the project folder; recover from the cloud Studio/R2/generating chat and save as Affirmology_WealthReading_Transcript_{Jeff,Sol}_v1.md (style exemplars only, never content-shared). ROLLOUT (Jeff): no more Mirror sends until rebuilt; beta signups open July 4 (framed as signup open, not delay), a couple audios every few days, reading cadence starts ~Fri Jul 10; beta testers must taste good ones first.
• INVESTOR PDF v2 (PHOTO + QUOTE PASS) SHIPPED (2026-07-02, Cowork): Affirmology_Mission_and_Opportunity_v1.pdf md5 70bc49c1..., 52pp, approved by Jeff for the site (Code re-uploads to media.affirmology.ai/docs/mission-and-opportunities.pdf). Added since v1: owner/"join a movement" section on the Ask (p49, glamorous Jeff+Sol photo); AUREA arms-raised crowd on the "Become part of this" finale (p52); founder/community photos (Jeff-in-the-circle p6, community-healing embrace p11 brightened, sacred-union candlelit circle p31, desktop-screenshot p29 with caption + bridge caption removed); 6 attributed pull-quotes (Jung p9, Robbins p14, bell hooks p21, Plato p25, Angelou p37, Durant p40); p33 fleshed-out italic caption (shared music-bed/start, individual hero's/Gene-Keys journey). NOTE: the editable HTML pipeline (/tmp/build.html + per-section renders) was LOST in a sandbox reset; this pass was done by compositing onto the rasterized live PDF (rebuild via pdftoppm of _v1.pdf -> PIL overlays -> re-save). To re-edit text/layout structurally, the HTML (Affirmology_Mission_and_Opportunity_v1.html, pre-2026-06-29-polish) must be reconstructed. Quote/photo bank + per-page ideas: Affirmology_BlankSpaceFillKit_v1.md.
• THE MIRROR + WELCOME VIDEO PUSH (2026-07-01, Cowork). Two new pieces going to the demo-takers: (1) INLINE WELCOME VIDEO on the results page: Jeff+Sol's ~3:43 talk media.affirmology.ai/experience/jeff-sol-welcome.mp4 (poster media.affirmology.ai/brand/jeff-sol-welcome-poster.jpg, the heart-hands frame). Cowork embedded it in result.html REPLACING the old jeff-and-sol photo, in the warm-note block right before the Inner Orbit / investor / beta asks (the video itself covers investment/beta/roles, so it leads into them). Under it, a one-line text link to the longer ~15-min YouTube talk https://youtu.be/koDNSAu6aA8 (page only; NOT embedded, NOT in the email). NOT yet pushed to git. (2) THE MIRROR email drafted + test-sent to Jeff (Affirmology_MirrorEmail_v1.html, subject "A new audio gift awaits"): announces The Mirror (a ~9-11 min deeper chart reading, spoken, NO music bed, like a paid astrologer/HD/Gene Keys reading; positioned as a small sample of the coming reading feature), drives to the results page (no MP3 attached), and teases the welcome video with its still. DEPENDENCIES before real send: Code must (a) generate each person's Mirror MP3 (Fish "Dr. Young" male voice) and expose it on their status payload, and (b) Cowork adds a "New Audio - The Mirror" section high on the results page under the Origin Soul Song. Deliberately does NOT link the 15-min YouTube in the email (Jeff's call: keep it driving to the page). Locked email templates so far: Welcome (Affirmology_WelcomeEmail_Template_v1.html), FormStopper (Affirmology_FormStopperEmail_v1.html), RenderFix Oops+Welcome (Affirmology_RenderFix_OopsWelcome_v1.html), Mirror (Affirmology_MirrorEmail_v1.html). BETA ROLLOUT STRATEGY (Jeff mulling July 4th app-readiness anxiety): advised DECOUPLE the date from the App Store, make 7/4 the COMMUNITY launch (open the chat/WhatsApp, founding welcome, manual nightly audio), let the app trail by days. Sequence: Mirror email today, then a beta "here's what to expect, click if you're in" opt-in email in a couple days, then open the community on/around 7/4 with 2-3 named social gatherers. Don't over-email.
• LAUNCH-DAY DIRECTIVES + LIVE BUG (2026-06-30, Cowork, Jeff's explicit calls): (1) KILL THE PER-RENDER PDF, HARD RULE. No PDF of any kind may be generated when a person does an Origin Soul Song demo render. The legacy/first demo generated a PDF that cost ~$0.45 every single render, a cost leak that must die NOW. The real report is a SEPARATE, single, polished PDF generated a DAY OR TWO LATER, and only after the new report process is solid, optimized, and fixed, NEVER per origin-song render and never inline with it. Code must strip PDF generation out of the demo/origin render path entirely. Future PDF runs on its own next-day job once nailed. (2) WELCOME EMAIL RENAMED: subject is now "Welcome to Affirmology" (was "Your Origin Soul Song is ready"); Code changed it. Better for everyone, especially people who got it a day late. Copy/subject will be revisited when the next audio + the real PDF go out. (3) BETA OPT-IN TRACKING BUG (launch-critical, OPEN, Code's lane): the results-page "I'm in" beta checkbox is NOT recording opt-ins. Evidence gathered: the LIVE result.html does POST to https://affirmology-studio-api.onrender.com/api/beta-signup, and that endpoint + CORS work from curl (my two diagnostic POSTs created records DIAG + CORSDIAG in People with "Beta committed" checked). But a REAL browser click (verified via Chrome on two live results pages) logs NOTHING, and the only "Beta committed" records in Airtable are my two curl diags. Leading hypothesis: on the results/hub page the client does not know the user's email (it defaults to "your inbox"), and the endpoint likely keys on / validates email and silently drops the junk one (returns 200, writes nothing) while the page shows "thanks" anyway (the fetch has a silent .catch). FIX PATH for Code: resolve the person server-side by token (= runId = job_id, which IS known and valid on the page) and set Beta committed from that; add server-side logging; verify with a real browser click landing in Airtable. PAST opt-ins (Marcie et al.) are unrecoverable server-side (only in each user's localStorage) so plan to re-ask. The investor box is TEXT-ONLY (eyebrow "Investor Opportunities" + heading "Own a piece of Affirmology" + copy), NO investor checkbox, this is correct and intended (it only mirrors the beta box's look). NEXT (Jeff, gated on the beta form being fixed): a week of nurture to gather a tribe, staggered emails, feedback forms, plus texts / voice notes / phone calls to everyone who took the demo.
• INVESTOR PDF SHIPPED + LIVE (2026-06-30, Cowork): "The Affirmology Mission & Opportunity," 52 pp, image-composed from Affirmology_Mission_and_Opportunity_v1.html via the PIL compositor (_compose_pdf.py). Deliverable Affirmology_Mission_and_Opportunity_v1.pdf (== _WIP.pdf, md5 976b1278...) is hosted by the Code chat at media.affirmology.ai/docs/mission-and-opportunities.pdf (overwrite that one key to update; gold card + inline embed on the results page). Composer now keeps figures/boxes/h3-sub-heads atomic (no clipped lines, no stranded headings, no bleeds), chart binds to its $650k title, JP Morgan quote pinned p3, drop cap = raised initial, Go-to-Market starts its own page, Jeff's bridge + Systems-to-Audio images embedded, Jeff's desktop screenshot on p29 with caption. OPEN (Jeff doing later): fill ~9 dead-space pages -> see Affirmology_BlankSpaceFillKit_v1.md (quote bank + per-page ideas + photo/text-box pattern). Checklist on p43-44 left split on purpose (Jeff: fine as is).
• NAMING (2026-06-26, Jeff): "MAYA" = the CONSUMER / end-user beta app (the affirmology-app repo, bundle ai.affirmology.app, the July beta listening app). When Jeff says "the Maya app" he means the consumer app, NOT Atlas (the creator studio, affirmology-atlas). All chats/Code sessions: treat Maya = consumer app.
• BRAND LOGO LOCK (SOURCE OF TRUTH FOR ALL LOGOS + FONTS: Brand/Affirmology_BrandKit_v1.md - read it before placing any logo or setting any type. It lists every locked logo file, the pre-cut icon/favicon sizes, the transparent cutouts, the palette, and the typography (wordmark = Cinzel/Trajan caps; report + product body = Cormorant Garamond + Inter + JetBrains Mono mono-eyebrows). Background/history: Brand/Affirmology_LogoDecision_v1.md, locked 2026-06-27.). THE Affirmology brand mark is the "ORBIT STAR": a gold faceted compass-rose star (the person) inside a gold circle crossed by a tilted elliptical orbit carrying gold orb(s) (the cosmic blueprint), on deep emerald. RESPONSIVE SYSTEM: the TWO-ORB version Brand/Affirmology Logo Two Orb.png is the PRIMARY/hero mark (deck, web, marketing, large icon, PDF cover); the ONE-ORB version Brand/Affirmology Logo One Orb.png is the small/icon variant and the chosen APP ICON (this is the real logo now on the Maya app icon affirmology-app/assets/icon.png and the Maya player assets/logo_mark.png). Atlas has its OWN sister mark, the glowing golden orrery/armillary sphere Brand/Atlas Logo.png (NOT the Orbit Star). Palette: emerald #0a2014/#082c18, gold #c89b3c/#d9b15a, bright #ecd9a8, on the deeper emerald radial bg #0e3a24 -> #0a2014. Wordmark: AFFIRMOLOGY in Cinzel (Trajan caps) with tagline "Subconscious operating system". DO NOT use Affirmology_Logo_SealMark_v1.svg or Affirmology_Logo_CelestialA_v1.svg (hand-coded PLACEHOLDERS, NOT the brand mark; the v2.1 PDF brief wrongly named the SealMark and it shipped on a draft cover before Jeff caught it). TRANSPARENT VERSIONS EXIST - use them on any non-emerald background (light pages, photos, video overlays): Brand/transparent/Affirmology Logo Two Orb Transparent.png, Brand/transparent/Affirmology Logo One Orb Transparent.png, Brand/transparent/atlas-orrery-transparent.png (2560px, gold marks on full alpha). The full emerald-disc PNGs stay the default for app icons + emerald layouts. (App-side transparent copies also live at affirmology-studio/affirmology-agent/templates/assets/logo_hero.png two-orb + logo_mark.png one-orb + affirmology-app/assets/logo_mark.png.) FULL BRAND KIT (2026-06-28): Brand/Affirmology_BrandKit_v1.md is the single source of truth for logos + fonts (locked files, pre-cut sizes, transparent cutouts, palette, typography, usage). ⛔ HARD RULE FOR ALL CHATS / CODE / COWORK / PDF + DECK BUILDERS: the logo is a Higgsfield-made RASTER PNG - embed the actual file (Brand/Affirmology Logo Two Orb.png primary, Brand/Affirmology Logo One Orb.png icon, Brand/Atlas Logo.png for Atlas, or the pre-cut sizes in Brand/icons/). NEVER recreate/redraw/"vectorize" the star-orbit mark in SVG/CSS/React/canvas/matplotlib, never ask an image model to "make the logo", and NEVER inline Affirmology_Logo_SealMark_v1.svg / Affirmology_Logo_CelestialA_v1.svg (or any hand-coded vector) in any document, PDF, deck, email, or web page. The PDF report template currently inlines the SealMark SVG - that is WRONG and must be swapped to the real two-orb/one-orb PNG. If the real PNG is not reachable, STOP and ask; do not improvise a vector stand-in. (The wordmark TYPE lockup is the only legitimately vector asset; the mark itself is never vectorized.)
• VISUAL ASSETS INDEX (ALL CHATS READ THIS - where every oracle photo, wisdom-council bust, and brand graphic lives, 2026-06-28). SINGLE SOURCE OF TRUTH: Affirmology_GraphicsLibrary_Handoff_v1.md (project root) - a copy-paste directory of every usable graphic with exact paths; paste it into any other chat (PDF/investor/video) so they can find and use the art. What it points to: (1) ORACLE + WISDOM HEADSHOTS - individual circular busts for all 17 core council + 14 Wisdom Circle in Council/Roster/cast/ (e.g. sophia.png, prometheus.png, rumi.png, tao.png); full-body character sheets in Council/OracleSheets/; locked Atlas model sheet Council/OracleSheets/Atlas sheet.png. (2) COUNCIL DISPLAYS - Council/Roster/Affirmology_CouncilConstellation_v1.png (busts in a cosmic constellation ring, the hero display) + flat boards Affirmology_CouncilRoster_16x9_v2.png / Affirmology_WisdomCircleRoster_16x9_v2.png. (3) ORACLE-IN-ROOM STILLS - Council/Rooms/ (council chamber, Atlas+Sophia, Athena+Prometheus glyphs, Sophia caring, etc.); the locked Olympus rooms in Council/OracleSheets/Olympus Rooms/. (4) WHITE-PAPER + COVER PLATES - Brand/photos/plate-*.png (all 16 white-paper plates + future vertical) and Brand/photos/cover-a-soul-song.png / cover-b-system-soul.png / cover-c-astrolabe-heart.png + concept plates (plate-cover-cosmic-body, plate-heart-coherence-listening, plate-directions-constellation). (5) LOGOS + FONTS - Brand/Affirmology_BrandKit_v1.md. DEEPER CHARACTER RECORDS (with Higgsfield CDN links, from other chats): CLAUDE OUTPUTS/Affirmology/Affirmology_CHARACTERS_MasterIndex_v1.md + Affirmology_WisdomCircle_AppHandoff_v1.md + Affirmology_FullCast_MasterSheet_v1.html. Generation queue + house style block for making more in the same look: Affirmology_Higgsfield_GenerationQueue_v1.md + Affirmology_OracleImageLibrary_ShotList_v1.md. RULE: when any chat creates new brand graphics, save them to the right folder above and add them to Affirmology_GraphicsLibrary_Handoff_v1.md so this index stays the one place everyone trusts.
• PDF REPORT v2.1 (ORIGIN BLUEPRINT) DEEP REDESIGN BUILT; 4 REPORTS EMAILED TO JEFF FOR REVIEW (2026-06-27, Claude Code). Per Affirmology_PDFReportV2.1_RevisionBrief_v1.md (authoritative; supersedes v2 on depth + design). Drove from Jeff's v2 verdict: "incredibly brief on each section" + "extremely ugly". Fixed depth, persuasion, AND design together. CONTENT DEPTH: deepened pdf_reading_council.py CHAPTER_SYSTEM (explanatory, glosses jargon in plain language before use, anti-presumption "you may have noticed" never "you were never broken", required takeaway + explainer fields, varied closings, ~240-360 words/chapter); REVIEW_SYSTEM now also rejects presumption, undefined jargon, missing takeaway (it caught real presumption on a Jeff chapter and forced a rewrite). FRAMING rewritten to a large multi-section Opus payload (max_tokens 8000, with a 3-retry + control-char-repair JSON guard because Opus occasionally emits a stray char in 8k tokens): cover_hook, invitation, multi-page method (what_blueprint_is / the_science = predictive brain + ~83% affirmations-fail + heart coherence ~6/min + theta + 7-21 nights / the_bridge / market_proof ~80% Gen Z ~72% / the_turn), 5 system_primers each with its Oracle, differentiator_line (the SYNTHESIS of all five is the wedge), quadrant_reveals, convergence across all five, soul_song_bridge, path_forward (Identify/Align/Transcend with the science, app_pitch with audio contexts, membership Relationships/Wealth Codes/In-Season/Master Reading, oracles-as-feature, teaser cards). DESIGN: new templates/sacred_audio_report.html embeds Cormorant Garamond + Inter + JetBrains Mono (templates/assets/fonts/, copied from the app node_modules), bakes texture (emerald gradient + deterministic base64 star-field + gold rules), uses the REAL assets (Affirmology_Logo_SealMark_v1.svg seal inlined + breathing spark, marble-and-gold Oracle avatars templates/assets/oracles/*.png), two registers (emerald cinematic marketing/chart pages + cream book-page reading register), gold-thread cards, mono eyebrows. NAMED-PAGE FIX (critical): @page carries the emerald bg + @page cream the paper bg, sections use break-before: page NOT empty spacer divs (which had created phantom blank pages), so every page fills edge to edge with NO blank/near-blank pages. QUADRANT BLUEPRINT PAGE: new agents/pdf_chart_svg.py renders our own branded SVGs from the chart (astrology wheel by longitude, schematic bodygraph defined-centers-gold + channels, Gene Keys four-gift spheres, Vedic South Indian square by rashi, numerology diamond, plus a five-system synthesis teaching diagram); page = 2x2 quad grid (Western/Hermes, HD/Prometheus, GeneKeys/Athena, Vedic/Agastya) + numerology diamond (Pythagoras), each with an Oracle avatar + "what this reveals", convergence band across all five. QC: scripts/pdf_qc.py now 9 gates = v2's five (schema, chart-fact incl. invented-cross-name, style, recognition, render band 12-28) PLUS v2.1's four (anti_brevity: chapters 150+ words and method 300+; anti_presumption: regex of presumptive phrasings; explainer: jargon-bearing chapters need an explainer; design_density: fonts embedded + Oracle art + SVGs + logo + hero frame + membership/reveals present). Title "Your Origin Blueprint" everywhere incl. filenames. COHORT: all 4 SHIP, 19 pages each, ZERO near-empty pages, person-pure. Jeff emailed separately (Resend 25dfa7b7); Sol/Colin/Josh emailed individually (PDFs ~10MB each from the embedded oracle PNGs, so one-per-email; Resend 7a7d082e/f9f5f905/26612260) with Affirmology_PDFReportV2_1_BackendReport_v1.md attached to Sol's. Outputs at CLAUDE OUTPUTS/Affirmology/PDFReportV2_1_Tests/<person>/. RUN RECIPE unchanged from v2. ONLY Jeff's hero photo + closing image remain framed placeholders. FOLLOW-UPS (flagged, not blocking): (1) downscale embedded Oracle PNGs (1024 -> ~256px) so PDFs shrink from ~10MB and batch-email; (2) per-person MARKETING PROFILE workstream from the brief (Sun sign + element, HD type/authority for cadence, loudest themes for which teaser leads, store on the lead record, use from week one) = its own workstream; (3) next-morning DELIVERY wiring stays deferred (the OTHER chat's uncommitted pipeline.py/jobs.py). Code UNCOMMITTED (Jeff: "do NOT commit, still active redesign"); my files isolated (pdf_reading_council.py, pdf_chart_svg.py, sacred_audio_report.py, template+assets, pdf_qc.py, test_pdf_report_v2.py). This SUPERSEDES the v2 Phase 1 bullet below on depth + design.
• PDF REPORT v2 (ORIGIN BLUEPRINT) PHASE 1 BUILT; 4 TEST PDFs + BACKEND REPORT EMAILED TO JEFF FOR REVIEW (2026-06-27, Claude Code). Per Affirmology_PDFReportV2_BuildSpec_v1.md + _Architecture_v1.md + _AhHaMoves_Registry_v1.md. Canonical engine = the NESTED affirmology-studio/affirmology-agent (the top-level affirmology-agent is the stale copy). NEW reading-council pipeline src/affirmology/agents/pdf_reading_council.py: (1) salience_pass (Opus) ranks chart themes by HEAT = Loud x Juicy x Wanted, ENDURING + EMERGING layers ONLY, NO progressions/transits (those are the paid in-season teaser); _apply_caps keeps the thresholded set (cap 7, default 6, floor 5, NEVER padded), spillover themes become the locked teaser cards; (2) generate_chapter (Opus) writes each chapter from render_all_facts(chart) minus the timing block, grounded in corpus tier A/B records retrieved by tradition + element_key, using the Ah-Ha moves (pattern-first, name-the-shadow-then-gift, felt-sense noun, body/family/timing anchors, reframe); (3) recognition_review (Opus, second agent) cross-checks each chapter for recognition + chart-truth and sends it back for rewrite (default max_rewrites=2). build_reading orchestrates. The cross-check WORKS: it caught fabricated facts ('Cross of the Sphinx', 'Lilith on your rising') and precise HD channel errors ('Channel 27/50 connects Sacral to Spleen and does not touch the Throat') and forced rewrites. sacred_audio_report.py REWRITTEN to v2 (cover title "Your Origin Blueprint", env AFFIRMOLOGY_PDF_COVER_TITLE): consumes build_reading + adds Opus framing prose (Invitation / Why This Works / Soul Song bridge / Path Forward + teaser cards / Close); extract_blueprint_glance reads HD type/authority/profile/centers/channels/cross + Gene Keys gates + Life Path straight off the chart (the blueprint page is labeled PLACEHOLDERS in Phase 1); repair_style strips em dashes deterministically + rewrites banned corporate vocabulary in place (fixes the offending PART, not a full regen). generate_origin_report is the entry; generate_sacred_audio_report kept as a back-compat shim so the existing audio pipeline/CLI do not break. NEW template templates/sacred_audio_report.html = the 8-section flow in an emerald + gold keepsake style with [ HERO IMAGE ] / [ CLOSING IMAGE ] + chart-graphic placeholder slots. NEW QC scripts/pdf_qc.py (analog of audio_qc.py): gates = schema / chart-fact (gate.line citations + invented Incarnation-Cross-name) / style (zero em dashes + banned vocab) / recognition (every council review passed, no fabricated facts) / render (pypdf page count 8 to 20); run_pdf_qc returns SHIP or HOLD; cannot-fail = a HOLD is held for human review, never shipped. Tests tests/test_pdf_report_v2.py (10 pass) + full agent suite 55 pass, no regressions. COHORT RESULT, all 4 SHIP after QC: Jeff (12 pg, 6 ch, 2 teasers), Sol (13 pg, 6 ch, 3 teasers), Josh (6 ch), Colin (13 pg, 6 ch; Colin's first run was correctly HELD by the recognition gate on a forced Gene Keys convergence and an HD-wiring error, then SHIPPED on a max_rewrites=2 regen, so the default was bumped to 2). Readings are person-pure (Sol's 'The Seat of Disappearance' vs Jeff's 'The Mind That Turns Dark Matter Into Gold', zero content bleed). Output at CLAUDE OUTPUTS/Affirmology/PDFReportV2_Tests/<person>/ (PDF + backend/payload.json + qc.json). The 4 PDFs + CLAUDE OUTPUTS/Affirmology/Affirmology_PDFReportV2_BackendReport_v1.md (heat scores, chapters chosen, rewrites, QC per person, the logic) were EMAILED to jeff@jeffparker.love ONLY (Resend id e0db0487), never to Sol/Colin/Josh (Jeff: "for me to review, not shipped to them"). LOCAL RUN RECIPE: top-level venv python + PYTHONPATH=affirmology-studio/affirmology-agent/src + DYLD_FALLBACK_LIBRARY_PATH=/opt/homebrew/lib (installed pango + gdk-pixbuf for WeasyPrint) + AFFIRMOLOGY_CORPUS_DB pointed at a REAL corpus (pulled the 19 MB corpus/corpus.db from R2 to scratchpad; the repo default path is a 72 KB stub; the cloud uses /data/corpus.db) + ANTHROPIC_API_KEY (in affirmology-agent/.env). Model = Opus claude-opus-4-8 throughout; override via ANTHROPIC_PDF_READING_MODEL / ANTHROPIC_PDF_MODEL. DEFERRED to avoid colliding with the OTHER Code chat's uncommitted affirmology-studio edits (script_generator.py, pipeline.py, jobs.py, structures.py, cli.py): the next-morning DELIVERY wiring (attach the PDF to the email + the tokenized landing page) stays OUT of pipeline.py/jobs.py until that chat commits. PHASE 2 (Jeff's cover photo + closing image + the branded SVG bodygraph / wheel / Gene Keys / Vedic graphics into the placeholder slots) NOT started; Jeff is creating the imagery. Code is UNCOMMITTED (the studio tree is dirty from the other chat); my files are isolated (new pdf_reading_council.py / pdf_qc.py / test_pdf_report_v2.py + rewritten sacred_audio_report.py + template) and can be committed without touching theirs.
• MAYA PLAYER ROUND-1 + APP-WIDE BRAND DEPTH PASS SHIPPED; MAYA BUILD 10 IN EXTERNAL-REVIEW QUEUE (2026-06-27, Claude Code). Implemented CLAUDE OUTPUTS/Affirmology/Affirmology_Maya_ConsumerApp_Feedback_Round1.md on the consumer app (affirmology-app/app/player.tsx): (1) the COVER astrolabe is now an ANIMATED looping muted expo-video of assets/orrery-loop.mp4 (expo-av had broken the EAS cloud build with EXEventEmitter.h not found / could-not-build EXAV; expo-video is the EAS-safe swap); (2) the FULL transport (back-15 / play-pause / forward-15, seekable scrubber, time, repeat-with-5s-pause) renders from LOAD in a paused state, not only after first play; (3) astrolabe lifted OUT of the card onto the canvas so its deep-emerald edges blend; (4) track title -> upright Cormorant (headlinePlain); (5) one-orb assets/logo.png + AFFIRMOLOGY AI wordmark fill the dead top space above Back; (6) the gold AFFIRMOLOGY AI wordmark now anchors ALL Maya tabs like Atlas (app/(tabs)/_layout.tsx). APP-WIDE KEYBOARD: keyboardDismissMode="interactive" (swipe-down like Messages, tap-outside kept) on the shared Screen + Maya chat/feedback + Atlas oracle scrolls. BRAND DEPTH PASS via affirmology-brand skill, centralized in affirmology-shared (commit 51303f1, both apps inherit by submodule bump): three-tone ground bg #061b10 canvas / emerald #0a2014 inset / card #112e26 panels so panels lift instead of washing; Card gains a gold-soft hairline border + quiet shadow lift; new Divider + ScreenHeader (mono eyebrow + serif title + lede + gold hairline) anchor every screen; warmer muted #cdbfa3, brighter line #1e5238, added goldSoft #d9b15a + hairline rgba(217,177,90,0.18). Applied to Maya Home/Library/Oracle/Profile/Player + Atlas Create/People/Room. ATLAS home-screen LABEL set to "Atlas" (ios.infoPlist.CFBundleDisplayName, commit c5d3fc5) so it stops reading "Affirmology..." next to the Maya icon. BUILDS (EXPO_TOKEN auth, Jeff's token kept for session): Maya EAS prod build 9 then build 10 (build10 = depth pass, supersedes 9), both --auto-submit to App Store Connect app 6784908433; build 10 attached to the existing EXTERNAL group "Affirmology Beta" (public link https://testflight.apple.com/join/nDJacr85) but its beta-review submission is BLOCKED because old build 6 is still WAITING_FOR_REVIEW (Apple allows one build/train in beta review; the ASC API key got 403 cancelling build 6 = that needs Admin role, not the key's role). HANDS-OFF watcher running (scratchpad/asc.py watch via scratchpad/ascvenv, PyJWT/cryptography/requests; key YGGKMDU9H2 / issuer 19bd7795 / p8 at affirmology-app/asc-api-key.p8) retries every 5 min and auto-submits build 10 the moment the slot frees. Build 10 is ALREADY installable on Jeff's phone via TestFlight now (external review only gates OUTSIDE testers). AUTH FINDING: Maya is NOT hit by Atlas's studio-token P0 -- Maya calls only /api/app/* with the per-user magic-link bearer (shared api.ts setToken from auth.tsx), a different path from Atlas's studioApi/studio-token endpoints; Jeff already saw Maya home+oracle load in prod. COORDINATION: the Atlas Round-2 feedback (CLAUDE OUTPUTS/Affirmology/Affirmology_Atlas_Feedback_Round2.md) P0 studio-token is ALREADY fixed server-side (see next STATUS bullet, ATLAS_MAGIC_LINK=1); its remaining P1 Create-layout items are being worked by COWORK in the other terminal, so THIS Code session deliberately did NOT touch Atlas screens (avoid clobber). WHEN COWORK BUILDS ATLAS: pull the shared submodule to 51303f1 first so Atlas inherits the depth pass + keyboard fix (and carries the "Atlas" label c5d3fc5) and the two chats' shared edits don't diverge. Commits: shared 51303f1; Maya d8f9911 (polish 4-7) / a63d021 (depth pass) plus the earlier player/keyboard commits; Atlas 7a75336 (keyboard) / 5c38b5d (depth pass) / c5d3fc5 (label). Maya typecheck + Atlas typecheck both clean.
• ATLAS v7 P0 FIXED: "Invalid or missing studio token" on Listening Room / People / selectors (2026-06-27, Claude Code). Root cause: the Atlas prod build sends the user's MAGIC-LINK SESSION token (StudioTokenSync -> setStudioToken from useAuth) to the cloud studio's protected endpoints, but require_token (main.py) only accepts that session when ATLAS_MAGIC_LINK is truthy AND the token resolves to a circle member (jeff/sol/colin); ATLAS_MAGIC_LINK was NOT set on Render while STUDIO_REQUIRE_AUTH=true, so every app call 401'd. FIX (server-side only, NO token embedded in the app, NO app rebuild): set ATLAS_MAGIC_LINK=1 on the Render service via the API + redeploy. VERIFIED in prod with a real magic-link session (jeff's emailed code -> session token, code read from Gmail): /api/people 200 (Jeff/Sol/Colin/Josh), /api/categories 200 (26), /api/library 200 (29). The existing TestFlight build now works on relaunch. P1 (Create page layout: active-structure header, structure nav, drag-resizable controls, Hermes chat gap, direction-vs-Hermes clarity) is the Atlas Create screen affirmology-atlas/app/(tabs)/index.tsx -- IN PROGRESS next. Consumer app also shipped to TestFlight earlier (App Store Connect 6784908433).
• THE RECOGNITION (Origin A/B variant) BUILT + VERIFIED, and SOL'S LIBRARY EMAILED (2026-06-26, Claude Code). Engine: added SCRIPT_SYSTEM_PROMPT_ORIGIN_RECOGNITION (Origin formula, recognition maxed, becoming dialed to one quiet closing line) in the CANONICAL nested agent affirmology-studio/affirmology-agent/.../script_generator.py, selected ONLY when structure_id == "the_recognition" (Origin + flex paths byte-unchanged); threaded structure_id through the agent CLI --structure-id, studio pipeline.run_script, and jobs._run_script (job.category_id). Added the_recognition Structure in api/structures.py (Foundational, identity=10 transformation=3 target 3.5, empty directive). Verified: pytest tests/test_starter_palette.py 7 passed; get_structure('the_recognition') -> The Recognition 3.5 10 3 ''. jeff A/B SCRIPTS rendered at /tmp/recog_ab (both clean: name early, zero HD numbers, no em dashes; recognition closes present-tense, origin closes on becoming). GOTCHA: local studio default AGENT_ROOT is the STALE top-level affirmology-agent (NO Origin); the canonical engine is the nested affirmology-studio/affirmology-agent (what Docker/cloud ship). Render with PYTHONPATH=affirmology-studio/affirmology-agent/src + the top-level venv; the agent CLI does --script and --audio as SEPARATE calls (a single --script --audio returns after the script). SOL DELIVERY: her audios live on R2 media.affirmology.ai/library/ (found the lost ~70-min reading = sol_WealthReading_full_clean.mp3, 72 min). Pulled R2 creds from Render env (boto3), uploaded a new Charlotte Recognition + morning/wealth/sleep tracks, emailed solballard@gmail.com ONE library email (Resend id 407b7e61, bcc jeff) with 7 named+described links. OPEN: sol_OriginSoulSong.mp3 on R2 is an OLD 7-min render (Origin target is ~4 min); re-rendering a tight ~4-min Charlotte Origin to overwrite that key (then Cloudflare cache-purge so the already-sent link serves the fix). R2 cache note: a pre-upload 404 gets cached; bust with ?v= or purge.
• ATLAS ROUND-1 P0s SHIPPED + STUDIO DEPLOYED (2026-06-26, Claude Code on the Mac). Native Atlas (affirmology-atlas, commits 29ab34b..c1f572a, typecheck-clean, JS-only so they fast-refresh in the Simulator, no native rebuild): (1) brand wordmark in the header; (2) the in-app PLAYER fixed (it was silently swallowing failures) = a persistent mini-player that rides ABOVE the tabs on every screen (play/pause + progress + buffering), an AUTO-REPLAY toggle with a ~4s breath between loops, and surfaced errors; background playback on a real phone is built-in but only provable on-device (Jeff is on the iOS Simulator, which does not do background audio reliably); (3) CREATION IS NOW EXPLICIT: talking to Hermes never creates anything, confirm dialogs gate compose AND render (render warns about ElevenLabs credits), an always-on "Talking about for " banner, Hermes grounded in the active structure so it stops saying "nothing shared," a note pops in on structure change; (4) reversible "Remove from room" clean-out. STUDIO BACKEND DEPLOYED to prod (pushed origin/main 3079a8d..f09ef53, Render auto-deploy verified live): three_act_arc locked template added (Naming/Embodying/Becoming, 8 min, deep, INTERNAL Studio template only, NOT on the consumer palette; per Affirmology_ThreeActArc_StudioSpec_v1.md; 16 studio tests green; NO render done yet, gated on Jeff + credits); plus a hide/unhide endpoint POST /api/jobs/{id}/hide + hidden column for the Listening Room clean-out. Cleaned the room (33->29): hid the Cutover Test + Jeff's 3 older Origin duplicates, KEPT Jeff's newest Origin + the demo candidate + Colin's + Sol's (reversible via unhide). Round-1 feedback tracker: Affirmology_Atlas_Feedback_Round1.md (P0s A/B/G + wordmark + clean-out DONE; P1s C structure-detail/dials, D create-your-own, E richer feedback incl VOICE NOTES, F push-link-tracking REMAIN). ElevenLabs: Creator plan tapped out (~121k chars used, resets Mon Jun 29 ~13:30 UTC), Jeff added $10 overage to bridge + is applying for the 1-year ElevenLabs grant; Soul Songs need Charlotte (ElevenLabs) now. NOTE: VOICE-NOTING is NOT in any build yet (chat is text-only); Jeff is scoping adding voice notes into Hermes for both apps.
• ATLAS LOGIN UNBLOCKED FOR LOCAL TESTING (2026-06-26, Claude Code on the Mac). Root cause of "magic-link 6-digit code never works": request and verify were hitting DIFFERENT backends. The Atlas app defaults its API base to https://studio.affirmology.ai (app.json extra.apiBaseUrl), but that pretty domain is behind Cloudflare Access and returns a 302 HTML login wall to any non-browser/headless call (/api/* -> 302), so the app's auth never reaches the studio cleanly; a code emailed by one backend (cloud Postgres) can't verify against another (local sqlite affirmology-studio/data/consumer.db), and vice-versa. The LOCAL backend itself is healthy: jeff@jeffparker.love is invited + a founding person in consumer.db, RESEND is configured in affirmology-studio/.env, and a mint->verify round-trip over http://127.0.0.1:8765 returns ok:true (leading-zero codes included). FIXES: (1) added a dev-only echo to affirmology-studio/api/consumer.py auth_request -- when AUTH_DEV_ECHO is truthy it prints the code to the server console AND returns dev_code in the JSON (gated off by default; MUST stay unset on Render so beta testers always get email and codes never leak). (2) affirmology-atlas/.env (gitignored) now sets EXPO_PUBLIC_API_BASE_URL=http://192.168.1.153:8765 so the app talks to the laptop studio; app.json production default is untouched. The local studio was restarted on 0.0.0.0:8765 (agent venv python) with AUTH_DEV_ECHO=1, logs at /tmp/affirmology-studio-dev.log. PRODUCTION PATH NOW WIRED (2026-06-26, same session): (a) affirmology-atlas/app.json extra.apiBaseUrl changed from the Access-walled studio.affirmology.ai to the raw headless https://affirmology-studio-api.onrender.com; (b) pushed affirmology-studio commit 3079a8d (dev-echo + new endpoint) to GitHub main -> Render auto-deployed LIVE and verified (/api/admin/invite no-token=401, with team token works; /api/auth/* healthy); (c) confirmed via Render API that the cloud has RESEND_API_KEY+FROM_EMAIL set and AUTH_DEV_ECHO UNSET (codes only emailed in prod, never returned over the wire); (d) cloud auto-seeds founding (jeff/sol/colin) on boot. NEW TOOL: POST /api/admin/invite (Studio team token, body {"emails":[...]}, idempotent) onboards beta testers WITHOUT a redeploy -- the cloud uses sqlite on the Render disk (STUDIO_DATA_DIR=/data, no Postgres), so this endpoint (or consumer_seed.py) is the only way to add invites. STILL OPEN: real beta-tester emails are not yet seeded -- Affirmology_Airtable_BetaTesters_v1.csv holds only placeholders (no addresses); waiting on Jeff for the real list, then one /api/admin/invite call adds them. Side note: the push also carried a previously-unpushed local commit ec2be78 (natal_sacred kill / Origin canonical) to the cloud, consistent with the Origin direction above. CONSUMER APP (affirmology-app) ALSO WIRED FOR LOCAL SIM TESTING (2026-06-26): affirmology-app/.env.local (gitignored) sets EXPO_PUBLIC_API_BASE_URL=http://192.168.1.153:8765; screens are wired (home->api.home, library->api.library, chat->api.chat, player->api.audio+api.feedback, login->useAuth/authRequest+Verify) and verified end-to-end against local (me/home/library/chat all 200; chat returns a real Sophia oracle reply in ~3s). jeff's local library was EMPTY (0 audios), so one test record was published to him via /api/app/publish (team token) pointing at an existing local render served at /media/... (ABSOLUTE url, since the player uses mp3_url raw and does NOT prepend base). CAVEAT: the player uses react-native-track-player (native module), so actual PLAYBACK needs a dev build, not Expo Go; home/library/chat/feedback/login are plain fetches and work in Expo Go. iOS DEV-BUILD PLAYBACK ENABLED (2026-06-26): affirmology-app/app.json ios.infoPlist.NSAppTransportSecurity pinned to {NSAllowsArbitraryLoads:false, NSAllowsLocalNetworking:true} -- the minimal ATS exception that lets the player stream the local HTTP media (http://192.168.1.153:8765/media/...) on a dev build; it ONLY permits local-network HTTP, does NOT weaken prod HTTPS (onrender), and Apple treats it as non-flagged. /ios is gitignored (managed prebuild/CNG) so app.json is the source of truth; the already-generated plist matches. Run the dev build with npx expo run:ios (Metro pins 8081 for dev-client). PORT HYGIENE (2026-06-26): affirmology-studio/start_public.sh now auto-frees the port before binding (macOS-safe, no xargs -r) and binds 0.0.0.0 (STUDIO_HOST override) so the apps' LAN base URL resolves, still pinned to 8765 (verified: started twice back-to-back with no manual kill -> auto-freed + rebound, single listener). Consumer /api/auth + /api/app are NOT behind require_token (host-based gate exempts localhost; team endpoints only), so start_public.sh (which sets STUDIO_REQUIRE_AUTH=true) is safe for the apps; prefix AUTH_DEV_ECHO=1 ./start_public.sh to keep the dev code-echo. Both apps' package.json start now FREES 8081 then pins it (P=$(lsof -ti:8081); [ -n "$P" ] && kill -9 $P; expo start --port 8081) -- because the NATIVE debug builds load JS from Metro at the fixed localhost:8081, so only one app's Metro can own 8081 at a time; running npm start in either app claims 8081 for that app (superseded the earlier --port 0, which wrongly pushed the beta Metro to 8082 where the native build never looks). CONSUMER APP "OPENS ATLAS" BUG SOLVED (2026-06-26): it was NOT a bundle-id or scheme collision -- installed IDs differ (ai.affirmology.app vs ai.affirmology.atlas), schemes were both absent (not colliding). Root cause: the installed ai.affirmology.app debug binary was STALE -- its native react-native-track-player did not match the current JS (Metro logged missing native sleep-timer methods), so it failed iOS preflight / crashed on launch; with the look-alike blank icon, opening it appeared to "fall back to Atlas." FIX: cd affirmology-app && npx expo run:ios rebuilt the binary (Build Succeeded); the run:ios auto-install had been interrupted so I simctl uninstall + simctl install the fresh Affirmology.app + relaunch -> boots the CONSUMER app, screenshot-verified showing Home "Welcome back, Jeff." / "Open your library" (tabs index/library/chat/profile), stable, logged in as Jeff. Icon to tap = "Affirmology" (blue chevron); "Affirmology Atlas" is the blank white icon. Schemes still absent on both (a future expo prebuild --clean would bake them; not needed for icon-launch).
• DEMO IS NOW CLOUD-NATIVE + the OLD natal engine is DEAD, replaced by the new ORIGIN SOUL SONG formula (2026-06-26, Claude Code on the Mac). The live demo runs entirely in the cloud with Jeff's laptop OFF: the Cloudflare worker affirmology-demo-worker now routes demo.affirmology.ai/api/* to a demo bridge added to the cloud Studio (affirmology-studio/api/demo_render.py: /render-sacred-audio + /status, token-gated by DEMO_AUTH_TOKEN), which reuses the Studio's raw-birth -> script -> audio+QC+PDF -> R2 pipeline. Verified live end-to-end (POST demo.affirmology.ai/api/render -> 202 -> rendered in cloud, R2-hosted). The agents (Hermes) + audio hosting (R2, media.affirmology.ai) are also cloud-native; Cowork now reaches Hermes at https://affirmology-studio-api.onrender.com/api/hermes (bearer STUDIO_AUTH_TOKEN); the Tailscale funnel is RETIRED (CLAUDE.md "Interfacing from Cowork" updated to the cloud endpoint). THE OLD NATAL-WALK SPINE IS KILLED (Jeff's explicit call, demo lock lifted): the engine's rigid SCRIPT_SYSTEM_PROMPT is dead code (no path selects it). The foundational/demo audio now renders SCRIPT_SYSTEM_PROMPT_ORIGIN -- identity-first opening ("I am [name]" by sentence 2-3), strong aspirational hooks, shadow->gift body with astrology placements KEPT (Virgo Sun in the 8th) but Human Design gate/channel NUMBERS OUT (translated to felt qualities), "I am the one who…" declarations + three-noun title, the REQUIRED hand-on-heart breath + embodied-gratitude/coherence beat, open-loop close (never a CTA), ~3-4 min, Charlotte, Heavenly Circuit bed, QC. natal_sacred relabeled "Origin Soul Song", target 3.0 min. Council bug fixed: pipeline.run_script only runs --council when a directive/concept is present, so the no-directive demo uses the Origin formula (it was silently going through the FLEX/jargon path before). Gold approved Jeff script saved: CLAUDE OUTPUTS/Affirmology/Affirmology_OriginSoulSong_Jeff_ApprovedScript_v1.md. System of record for audio structures: Affirmology_AudioStructures_Tracker_v1.csv (golden value + status per parameter) + Affirmology_AudioStructures_Formula_and_WIP_v1.md. STANDING RULE: every Origin audio email to Jeff includes the golden script + the new script + a written audit, and the ACTUAL rendered script is verified (name present in the first ~3 sentences, ZERO HD gate/channel numbers) before sending. Keys: Cloudflare demo token in affirmology-studio/.env as CLOUDFLARE_DEMO_TOKEN (DNS + worker edit); Render API key drives deploys; Zoho email DNS for affirmology.ai all live. OPEN/NEXT: lock the Origin formula by ear, wire the proven 0.96 speed into the demo render path (currently the slower ~0.95 standard pace -> ~4 min), trim ~60 words toward the ~430-word gold; then The Mirror (9-11 min reading, Fish male "Dr. Young") + the PDF rework; Fish Audio for scaled production (demo voice stays ElevenLabs/Charlotte). NOTE: this supersedes the now-stale "## 4. Demo audio baseline (locked)" and "## 5. Demo backend (laptop)" sections below.
• RESOURCE HUB BUILT (2026-06-26, Cowork, NO engine code touched). New INTERNAL gated site planned at resources.affirmology.ai for Jeff, Sol, Colin: a synthesized, CLUSTERED, searchable navigator over the project docs, NOT 421 flat pages. Per Jeff's steer the front door is curated topic CLUSTERS with written overviews (e.g. the many scattered investor-video idea docs collapse into ONE "Investor Video" cluster), plus a client-side search over everything so nothing is lost, and superseded versions auto-routed to Archive. Cowork built the whole generator + curated taxonomy + house-style CSS (extracted from Affirmology_LaunchCRM_Playbook_v1.html) + a rendered dist from 421 docs, all in resource-hub/. It indexes CLAUDE OUTPUTS/Affirmology IN PLACE (no files moved, the 2 hardcoded paths in hermes_tools.py + craft_seed.py untouched), so the FileOrg Phase 1/2 cleanup is now OPTIONAL and de-risked; the generated manifest is the ready catalog if a physical migration is ever wanted. NOW LIVE (2026-06-26, Claude Code on the Mac, deploy paste executed end to end): the hub is DEPLOYED + GATED at https://resources.affirmology.ai (Cloudflare Pages project affirmology-resources, 422 docs / 48 archived). The Cloudflare Access gate was put in FRONT before the custom domain: Access app "Affirmology Resources" with one Allow policy "Team allowlist" = exactly jeff@jeffparker.love + solballard@gmail.com + cjacobsandassoc@gmail.com (everyone else default-block); verified live (unauthenticated GET 302s to the Access login at weathered-feather-0bae.cloudflareaccess.com, no content served; dedicated TLS cert CN=resources.affirmology.ai active). The proxied CNAME resources -> affirmology-resources.pages.dev was created with CLOUDFLARE_DEMO_TOKEN (the wrangler OAuth + studio .env tokens lack DNS write; the Access app + custom domain were created with a one-off Access-scoped token Jeff pasted, since neither standing token has Zero Trust Access scope). The launchd 2-hour cron ai.affirmology.resourcehub is INSTALLED + loaded (RunAtLoad + every 7200s) and its regenerate->wrangler-deploy path is validated under a stripped launchd env; rebuild.sh was hardened to add Homebrew to PATH (launchd's minimal PATH would otherwise never find npx/wrangler). Incremental rebuild proven (edit 1 doc -> reprocessed 1, unchanged docs skipped). Cron relies on wrangler's stored OAuth login to deploy unattended (no .deploy_env needed; if Jeff ever logs wrangler out, the cron deploy step will fail until re-login OR a Pages-scoped token is dropped in resource-hub/.deploy_env). Paste-ready brief: Affirmology_ResourceHub_DeployPaste_v1.md. Generator + taxonomy in resource-hub/ (tune topics by editing resource-hub/clusters.json, no code). Original build brief: Affirmology_ResourceHub_BuildBrief_v1.md.
• DEMO LOCK LIFTED (2026-06-26, Jeff's explicit call, overrides the CLAUDE.md hard rule): the old "demo is locked, never touch it" rule is OFF. The demo is NOT live, nobody is using it, and it is being fully replaced by the new Origin Soul Song (origin_soul_song, optimized in a separate Code session) and new videos/graphics/investor pieces. natal_sacred is being KILLED everywhere (studio + palette + the demo render path if it uses it). It is fine to break the old demo. Everything is in draft mode. Keep changes committed so anything is recoverable from git. ALL CHATS/CODE SESSIONS: do not re-lock or protect the demo.
• REBOOT RESUME (Origin Soul Song session, 2026-06-26): safe to reboot, NO render in flight (v5 done, email sent). SURVIVES reboot (cloud/R2): the new Origin Soul Song renders + Listen links (media.affirmology.ai), the cloud Studio + the new Origin formula (on Render), Hermes/agents (cloud-native), Zoho/email DNS, Cloudflare token. BREAKS on reboot (laptop-served, both fine to leave down): the live demo (demo.affirmology.ai still routes to the laptop; cloud bridge built but NOT cut over, restore via laptop server + funnel OR do the cutover), and the OLD Fish-voice-test Listen links (Tailscale funnel; recent Origin links are R2 and keep working). WHEN BACK, pick up: (1) listen to the latest Origin Soul Song + direct refinements (formula now carries every beat); (2) wire 0.96 speed (~3:45) + optionally trim ~60 words; (3) CUT OVER the demo to cloud (one move, makes it survive reboots/laptop-off = the North Star); (4) retire Tailscale, point Cowork at the cloud Hermes endpoint; (5) open structures still to build: The Mirror (9-11 min) + the PDF rework. Persisted in memory + CLAUDE OUTPUTS/Affirmology/ (tracker CSV, formula doc, gold approved script).
• IDENTITY + TRANSFORMATION RESEARCH LANE (2026-06-26, Cowork; NO engine code beyond one craft row; brand/video CREATIVE is owned by Jeff's other master-doc/video chat, NOT here). Audited the corpus build: reading corpus healthy (25,570 records across 6 traditions) but the nightly launchd build has been IDLE since Jun 18 (one run stuck 'running'); the wisdom/practitioners research lane ran ONE batch Jun 21 (14 hypnotherapy craft entries) then stopped with 96 thinkers still queued = the identity/transformation canon (Maxwell Maltz, James Clear, Jung, IFS/Schwartz, Byron Katie, Neville, Coue, van der Kolk, Gabor Mate, etc.). KEY FINDING: the cloud version of the nightly job ALREADY EXISTS in code (affirmology-studio/api/growth_cron.py, written to run as a Render Cron Job "No laptop", never switched on); only the laptop launchd timer (com.affirmology.nightly.plist, meant for the un-set-up Mac mini) is live, which is why it stalls when the lid closes. JEFF GREENLIT moving the WHOLE corpus-wisdom build to the cloud (deploy growth_cron as a nightly Render Cron Job, push freshest corpus to R2 as canonical, retire the laptop launchd, load a WIDE identity roster not a few names + mine Substack as a Phase-2 source tier, identity-first, cost-capped). EXECUTED 2026-06-26 from Cowork via Render API + R2 (boto3), keys from affirmology-studio/.env + the Render service env. DISCOVERY: the cloud cron ALREADY EXISTED and was running nightly (Render cron affirmology-growth-nightly crn-d8quk8ugvqtc73e9na9g, 0 9 * * * UTC, R2 bucket affirmology-media key corpus/corpus.db); the "stalled" look was only because the LOCAL corpus.db Jeff saw was a stale Jun-21 snapshot while the cloud advanced daily; it only mined 6/night (default batch unset). DONE: (1) widened research_backlog +33 with a robust identity/transformation roster (Dan McAdams narrative identity, Bruce Ecker memory reconsolidation, Lisa Feldman Barrett + Andy Clark predictive brain, Rick Hanson, A.H. Almaas essence, Hal & Sidra Stone, Gendlin, Pat Ogden, Deb Dana, Janina Fisher, Fosha, Steven Hayes, Kristin Neff, Tony Robbins, Benjamin Hardy, Caroline Myss, Estes, Plotkin, etc.) + the "Is it true, or just familiar" Socratic craft line; (2) set GROWTH_PRACTITIONERS_BATCH=120 + GROWTH_PRACTITIONERS_MAX_COST_USD=5 on the cron; (3) triggered a manual run (job-d8vc83svikkc73ffgja0) that SUCCEEDED and cleared the whole backlog: research_backlog 143 done / 0 pending, practitioners craft 143, total craft 252; (4) DELETED the Mac's local corpus copies (corpus/corpus.db + 2 .bak) per Jeff = R2 is the SOLE source of record now (this intentionally breaks the laptop Hermes funnel's default local-corpus path so everything is forced onto the cloud; the cloud Studio/demo/render already read R2 and are unaffected). INTERFACE BUILT: Cowork scheduled task wisdom-corpus-daily-briefing (daily 08:02 local; /Users/jeffreyparker/Claude/Scheduled/wisdom-corpus-daily-briefing/SKILL.md) that pulls the corpus from R2 to /tmp (NEVER stores it on the Mac), writes a SYNTHESIS (what is useful + what NEW IDEAS it sparks, not a pull log), keeps its memory in R2 (reports/briefing_state.json), and is a chat Jeff can reply to in order to probe the corpus + add research leads (appended to the R2 backlog for the next run). Substack = a BROAD Phase-2 source tier (not limited by topic/modality), not built yet. NEXT: build the Substack source tier; consider re-pointing the engine default corpus path / laptop funnel to R2-only; Jeff can hit "Run now" on the briefing task for the first synthesis immediately. Named our identity model = ESSENCE-CONGRUENCE (chart reveals essence, install congruence; wedge = others make you guess your essence, we read it, so the identity is already yours and wires in faster). SHARED RESEARCH DOCS at project root for OTHER chats to search/pull/synthesize (these are feeders for Jeff's master docs + video chat, not finished creative): Affirmology_IdentityResearch_AuditAndMap_v1.html (the audit + the beladivine "Identity Shifting 101" reference mapped to our craft corpus + gaps + the 96-name roster + captured frameworks), Affirmology_IdentityNeuroscience_BrandStory_Synthesis_v1.md (copy-ready 18-to-254-day essence-congruence argument in multiple registers), Affirmology_InvestorVideo_EducationalSegment_v1.md (extracted teach-then-reveal education for the vision/investor video), Affirmology_BlueprintAsCharacter_Model_v1.md (CANONICAL model tying identity + transformation: blueprint casts the character, soul plays it, design schedules the ego-deaths/rebirths via Gene Keys sequences + Saturn returns; personalization holds BOTH the designed character AND the lived soul-state, learned via the chatbot + questionnaire discovery layer). CONCEPTS CAPTURED for later build: the "Transformation Architect" internal design agent (holds craft + install-rules, tests journey sequencing WITH Jeff, separate from Hermes); a dedicated NARRATIVE IDENTITY journey (real gap, worth a flagship build); reframing the witnessing/relationship layer (Constellation, invites) as identity-SOLIDIFICATION. Living tracker = the AuditAndMap HTML.
• TRANSFORMATION STRUCTURES + TECHNIQUE PIPELINE PLANNED (2026-06-26, Cowork briefing chat; NO engine code touched, specs only). From the Wisdom Corpus briefing Jeff greenlit building two transformation structures NOW for testing and tracked a set of decisions. SPECS WRITTEN (paste-ready for Claude Code): Affirmology_ThreeActArc_StudioSpec_v1.md (new three_act_arc structure, Naming/Embodying/Becoming, 8 min, plugs into the existing directive->concept flex path in api/structures.py) and Affirmology_ReconsolidationBreakthrough_StudioSpec_v1.md (new reconsolidation_breakthrough structure: Permission Narrative open + memory-reconsolidation pairing + nightly incantation close, driven by a NEW private person-level working_through limiting-beliefs field populated by self-entry+chat+questionnaire). IDENTITY DELIVERABLE: Affirmology_IdentityResearch_v1.md = the 62-entry identity-install toolkit organized into a Tier-1 non-negotiable stack (self-relevance, recognition-over-instruction, I-am present tense, listener-language mirroring, prediction-error personalization) for the first-week "this gets me" hit, plus structure candidates. REGISTRY (the durable "list somewhere"): Affirmology_TechniqueToolkit_and_Ideas_Registry_v1.md holds (1) the 5-stage technique pipeline mined->candidate->council-proposed->studio-tested->approved-toolkit with ROTATION so audios don't repeat techniques, (2) the corpus browser as a TAB on resources.affirmology.ai (static-generated by the existing hub cron from R2, needs its own build brief), (3) the nightly cadence = nightly repetition within a monthly/new-moon focus cycle, (4) Hero's Transit (approved direction + transit-eligibility/timing model: cast IN-IT live, JUST-FINISHED as integration, APPROACHING as call-to-adventure, don't offer if no major transit near), (5) Possible Selves = PREMIUM logged, Songs-from-highlights + Sonic-Anchor TMR = experimental logged. NIGHTLY MODEL DECIDED (Jeff, 2026-06-26): the MONTHLY BAKED AUDIO model. Week 1 = identity-only nightly (Origin/The Recognition). After that, ONE baked sleep audio per month, listened nightly, LOCKED for the cycle, baked at the new moon; it grounds in identity (not a permanent base layer, just grounding + the Tier1 open), leans lightly on that month's moon/transits, and is shaped by the person's ONE current focus belief. NO extra nighttime bakes for anyone (premium unlocks OTHER content: constellations, journeys, depth pieces, NOT more nighttime audios). working_through field = EXACTLY ONE belief at a time (replaceable, not a list). Voice/tone shaping = FUTURE (no multiple voices yet). THIRD SPEC WRITTEN: Affirmology_TheRecognition_StudioSpec_v1.md (new the_recognition Foundational structure, A/B variant of Origin Soul Song; recommended impl = a SCRIPT_SYSTEM_PROMPT_ORIGIN_RECOGNITION engine sibling selected by id so the A/B isolates recognition-density vs becoming; Origin path byte-unchanged). DECISION on Origin: do NOT replace Origin yet; A/B The Recognition against it, the test decides replace/fold/separate; independently land Tier1 recognition principles across the first-week suite as craft. Your Wound Is Your Gift = premium depth piece (confirmed). Self-as-Context = a transformation TOOL folded into structures, not standalone. NEXT: hand the 3 specs to Claude Code (Three-Act, Reconsolidation, The Recognition); build the technique toolkit so available identity/transformation tools are browsable + sprinkleable; write the corpus-browser build brief (tab on resources.affirmology.ai).
• NATIVE BUILD PHASE 1 + 2 DONE (2026-06-26, Cowork; full log CLAUDE OUTPUTS/Affirmology/BUILD-PROGRESS.md). PHASE 1 (shared foundation): new repo affirmology-shared/ (commit 81f0baa) is the ONE source for both apps (design system, UI kit, magic-link auth, API client, oracle faces, player, event queue); the consumer app affirmology-app/ was refactored to consume it as a git submodule at ./shared (commit a9c92e9) with NO regression (tsc clean, identical iOS bundle). The submodule lives UNDER each app root because Metro's node crawler (no watchman on this Mac) only resolves files under the root AND EAS only uploads the project tree; a sibling package + watchFolders was proven not to work. PHASE 2 (native Atlas): new repo affirmology-atlas/ (Expo SDK 56, bundle ai.affirmology.atlas, commits c2f1d5c + 621cf87) on the shared foundation: magic-link Login, the HERMES CREATION centerpiece (chat-primary + reachable controls panel + council "working" theater + co-creation quick-replies, wired to /api/hermes + /api/compose + /api/jobs + /api/audio), People (data-driven charts), Listening Room (/api/library play + Love/Needs-work/Approve + push), and an A/B Compare view. Structures/people are FETCHED from the backend (data-driven, no app release to change them). Verified: tsc clean, iOS export, expo-doctor 21/21. BACKEND (affirmology-studio commit d08c75f, Studio-only, additive, demo byte-untouched, LOCAL not pushed): creator roles (api/roles.py + GET /api/roles + soft compose gate behind CREATOR_ROLES_ENFORCED, off by default) and A/B variant groups (variant_group/variant_label columns + /api/jobs/{id}/variant + /api/variants/{group}); tests test_roles.py + test_variants.py; full suite 172 passed, 1 failed = the SAME pre-existing test_chart_cache nondeterminism. NEXT (with Jeff): seed the named starter palette as data (a script-content change), run Atlas in the Simulator once the Xcode license is accepted, wire the per-audio metrics as creation controls, flip CREATOR_ROLES_ENFORCED when the opt-in map is right, push the studio backend when ready. Jeff steps in BUILD-PROGRESS.md "FOR JEFF".
• DEPLOYED (on origin/main, cloud auto-redeploys): INFRA TASK 2026-06-24 ALL FOUR DONE. (1) cloud render is LAPTOP-INDEPENDENT (proven, see below); worker secret STUDIO_FUNNEL_URL pinned to https://affirmology-studio-api.onrender.com (no laptop funnel). (2) media.affirmology.ai public serving LIVE (HTTP 200, R2_PUBLIC_BASE_URL set). (3) C21 Hermes chat = clickable links + inline audio/PDF player (25c15b4). (4) C23 EMAIL LIVE (see below). Also LISTENING ROOM shipped (6d233d2) = one Studio page where every audio lands, inline play + Download + Love/Needs-work/Approve, fed by new GET /api/library (API-first for the future iOS player). Prior app-changes all still shipped (C29 347bc09, C28 3975330, C26 601e7f0, C3 13becb2). Demo stays locked + byte-for-byte unchanged; C3 tones OFF by default.
• iOS APPS OVERNIGHT BUILD DONE (2026-06-25, Claude Code on the Mac, all three briefs built + tested + committed in small steps; nothing touched the locked demo; no em dashes). Read CLAUDE OUTPUTS/Affirmology/OVERNIGHT-REPORT.md + MORNING-FOR-JEFF.md. BRIEF A (commit cdfc834 + email guard c765e99): consumer system-of-record + magic-link auth added to the affirmology-studio API (api/db.py, db_models.py per spec s10, auth.py, consumer.py /api/auth/ + /api/app/, consumer_chat.py guardrailed oracle with NO audio-creation tools, consumer_seed.py). SQLAlchemy with Postgres via DATABASE_URL and a SQLite FALLBACK (runs now without a live DB); tests tests/test_consumer.py 12/12. BRIEF C (NEW repo affirmology-app/, commit 20edcd3): Expo SDK 56 + TS + expo-router consumer app, bundle ai.affirmology.app; login/home/library/player/chat/feedback/report/profile; react-native-track-player (bg+offline+lock-screen) behind one src/player.ts seam; offline event queue; 17 oracle photos in assets/oracles (current names incl Sri Agastya). Verified without Xcode: tsc clean, expo-doctor 21/21, full iOS Metro export succeeds. BRIEF B (commit 9cfd62a): Atlas Capacitor wrap (affirmology-studio/atlas-capacitor, ai.affirmology.atlas, remote server.url) + web/atlas-login.html magic-link sign in (additive) + require_token also accepts a circle member's session gated by ATLAS_MAGIC_LINK (OFF by default = live Studio unchanged); tests/test_atlas_auth.py 7/7. Full studio suite 138 pass / 1 pre-existing chart_cache failure (unrelated). BLOCKERS for Jeff (all in MORNING-FOR-JEFF.md, none blocked the night): (1) create Render Postgres + set DATABASE_URL (billable plan choice + live redeploy, not auto-done); (2) install full Xcode + CocoaPods (this Mac has only Command Line Tools, so no Simulator run / iOS build / pod install); (3) Apple: App Store Connect app records, an ASC API key for EAS, testers, an APNs key; (4) Cloudflare Access relax + set ATLAS_MAGIC_LINK=1 when moving Atlas to magic-link. NEXT: install Xcode+pods, then next session launch the app in the Simulator + verify end to end; create the Postgres; publish a few audios via POST /api/app/publish (team token) so the founders have content.
• LAPTOP-CLOSED RENDER PROOF (2026-06-24): a real Jeff / natal_sacred render ran entirely on the cloud (Render compute + R2 storage, cloud-token auth, laptop out of the loop), MP3 + PDF returned HTTP 200 via curl. MP3 https://media.affirmology.ai/a8e6c73f1ead46b193d03175787f62d3/Jeff_Parker_SacredAudio_Custom_v1.mp3 , PDF https://media.affirmology.ai/a8e6c73f1ead46b193d03175787f62d3/Jeff_Parker_SacredAudio_Custom_v1_Report.pdf . Health green, agent_python now resolves to /usr/local/bin/python (the old script-but-no-audio bug is gone).
• LAPTOP-CLOSED RENDER RE-PROVEN (2026-06-25, Cowork): re-ran a real Jeff / natal_sacred render end-to-end against the CLOUD (compose -> script_ready -> audio -> done, all on affirmology-studio-api.onrender.com, R2 storage, cloud-token auth; laptop only polled). Job 61a13c6890384668b56b5a1653d89e1d. MP3 https://media.affirmology.ai/61a13c6890384668b56b5a1653d89e1d/Jeff_Parker_SacredAudio_Custom_v1.mp3 (HTTP/2 200, audio/mpeg, 12,449,166 bytes, ID3 header) , PDF https://media.affirmology.ai/61a13c6890384668b56b5a1653d89e1d/Jeff_Parker_SacredAudio_Custom_v1_Report.pdf (HTTP/2 200, application/pdf, 42,765 bytes, %PDF). Whole render ~5 min. Confirms Listening Room (6d233d2) + C21 (25c15b4) are LIVE on origin/main + the cloud (GET /api/library HTTP 200, 25 entries). NOTE: studio repo is locally AHEAD of origin/main by 3 UNRELATED commits = overnight Brief A/B app build (cdfc834 Brief A consumer auth/db touching api/main.py, c765e99 email kill-switch, 9cfd62a Brief B Atlas wrap), NOT pushed (would redeploy prod), awaiting Jeff's go.
• DEV-ENV NOTE (updated 2026-06-26): Xcode license is now ACCEPTED (Jeff did it). Plain git works again (Xcode-bundled git 2.50.1); the old /Library/Developer/CommandLineTools/usr/bin/git workaround is no longer needed. Xcode 26.5 selected at /Applications/Xcode.app, xcodebuild -version clean. REMAINING gates before an actual iOS Simulator run of the RN apps: (1) CocoaPods is NOT installed (pod not found) needed for the native dev build (npx expo run:ios runs pod install) -> sudo gem install cocoapods or brew install cocoapods; (2) no iOS Simulator runtime appears installed -> open Xcode once and install an iOS platform/runtime (Xcode > Settings > Components), or xcodebuild -downloadPlatform iOS. Until then verification of the apps stays at tsc + expo export (both green).
• BRIEF A/B PUSHED TO PROD (2026-06-26, Jeff "push it"): the 3 overnight app-build commits (cdfc834 Brief A consumer system-of-record + magic-link auth, c765e99 email kill-switch, 9cfd62a Brief B Atlas wrap) pushed to origin/main; Render auto-deploy dep-d8v93gmrnols73c0ufi0 went LIVE and prod is healthy (status ok, /api/library HTTP 200, R2). SQLite fallback still in use (DATABASE_URL/Postgres not yet set). Studio repo now in sync with origin/main.
• C31 NEGATION RULE BUILT (2026-06-26, Studio-only, demo byte-for-byte locked, NOT committed, no paid audio). (a) NEGATION_RULE Orpheus draft directive injected only on the council path + a deterministic Apollo negation_gate() in agents/pacing.py wired into compose() beside pacing_gate (present-tense negation of an unwanted state in the body fails -> note back to Orpheus); Apollo's craft-gate prose also gained the negation/named-unwanted-state flag (5-gate JSON shape unchanged). (b) new corpus/craft_negation_seed.py = idempotent no-API seeder for the Chiron+Persephone two-pole (repressive+reactive) shadow-reroute entry (area breakthrough). (c) _negation_directive(pace): two-pole reroute for deep/standard, BUILD-ONLY for light activation/daytime. Tests tests/test_c31_negation.py (12). Spec: Affirmology_NegationRule_Craft_v1.md.
• C30 / OPTION B LAYERED STEM BED ENGINE BUILT (2026-06-26, Studio-only, demo locked, additive + OFF by default, NOT committed, no paid audio). New agents/stems.py (pure: palette select by bed_profile, manifest IO, per-pace bed arc + gain automation, key->tonal-center Hz, section timing). audio_mix.py gains assemble_stem_bed() (loop/stretch layers to section markers + length, per-section gain automation, loudnorm to tagged LUFS, sum) and apply_layered_bed() (select palette -> assemble -> duck under voice -> C3 tone tuned to palette key -> run audio_qc on the FINAL mix); graceful fallback to the single preset bed (or dry voice) when no palette. apply_tone_layer() gained tonal_center_hz (default 136.1 = C3 unchanged). stems/ library docs + MANIFEST_SAMPLE.json (no binaries; real layers come from Suno per the spec, AFFIRMOLOGY_STEMS_DIR env). The 5 acceptance tests + pure-config tests in tests/test_c30_stems.py (12, synth placeholder layers, no Suno/no ElevenLabs). NOT yet wired into the live Studio render path (additive only), so nothing changes until a palette exists + the call is wired. Spec: Affirmology_LayeredStemEngine_OptionB_Spec_v1.md.
• TEST STATUS (2026-06-26): engine cd affirmology-agent && .venv/bin/pytest = 70 passed. Studio cd affirmology-studio && ../affirmology-agent/.venv/bin/python -m pytest = 162 passed, 1 failed = the SAME pre-existing test_chart_cache::test_determinism_cached_vs_fresh (documented aspect-ordering nondeterminism, unrelated). All C30/C31 engine changes vendored byte-identical into affirmology-studio/affirmology-agent. No em dashes. NOTE: run the studio suite with python -m pytest (not the pytest entrypoint) so the studio root is on sys.path for import api.
• EMAIL LIVE (C23, 2026-06-24, commit 1dca6da): Resend is wired and proven. affirmology.ai is verified in Resend; RESEND_API_KEY + FROM_EMAIL=reports@affirmology.ai set on Render; shared api/emailer.py sends with a browser User-Agent (gets past Resend's Cloudflare error-1010 bot rule). EVERY finished render now auto-emails a notification (for-whom, audio title, who generated it, why/brief, when, plus a Listen button + MP3 download + PDF link) to NOTIFY_EMAILS (default jeff@jeffparker.love), wrapped so a mail failure never breaks a render; /api/send-digest refactored onto the same sender. Proven: 3 test sends + 1 real render auto-email all HTTP 200, Jeff confirmed receipt in inbox. The reason no email ever arrived before: Resend was never connected (no key, no from-address). POLISH (c92fe75 + 818c7b0 + 28ea47b): the render email goes to the person it is FOR, greeted by first name, in a branded emerald/gold template (Listen button + MP3 + PDF) with a no-reply footer. Recipient emails set in api/people.py: jeff@jeffparker.love, sol=solballard@gmail.com, colin=cjacobsandassoc@gmail.com (josh still unset, falls back to owner). NO personal-inbox BCC: an archive copy of EVERY render BCCs RENDER_BCC=jalanparker@gmail.com (set on Render); Jeff to add a Gmail filter on from:reports@affirmology.ai to skip-inbox + label "Affirmology Renders". (Optional next, not done: Cloudflare Email Routing so replies to reports@affirmology.ai forward somewhere.)
• PENDING JEFF'S CALL: (1) C3 tone listen-check still open (tones OFF until then); (2) nightly practitioners backlog growing. No build is blocked on me.
• iOS APPS / TESTFLIGHT (2026-06-24, Cowork, ARCHITECTURE DECIDED with Jeff; pending his final go on bundle IDs + greenlight): Apple Developer account ACTIVE. SPLIT DECISION (corrects the earlier "wrap everything in Capacitor" plan): ATLAS (internal Studio) = WRAP the existing web app (Capacitor or installable web), do NOT rebuild native (no payoff for an internal tool; wrapping is not a lock-in since the engine lives server-side and the face can be swapped later); move Atlas to MAGIC-LINK login (away from Cloudflare Access) since more creators will join the Studio. AFFIRMOLOGY consumer app = build NATIVE via REACT NATIVE (Expo), NOT a web wrapper (Jeff strongly prefers native; it is media-first, needs offline+BACKGROUND audio, fast iteration, Android later, public App Store in Aug). RN nails background/offline audio (react-native-track-player) + OTA updates (push JS/UI changes with NO Apple review; only NEW NATIVE features need a build/review). CHANGE-CADENCE clarified for Jeff: content/audios are DATA (instant swap, no release, no review, either tech); UI/logic via OTA (minutes, no review); only new native features hit a ~1-day review. CORRECTED earlier bad info: since the 2025 Epic v Apple US ruling, US apps CAN link out to the website for payment (commission on external links still being litigated mid-2026); so most billing happens on the WEB pre-download and in-app upgrade link-outs are allowed in the US. v1 REFRAMED as a TESTING SUITE (not the final paid client): data-driven library, variable 5-10 audios swappable any time, mandatory in-app offline+background player (NO emailed MP3s to beta users), scoped oracle chatbot (no audio creation, oracle character + photos, optional Fish voice), feedback+ratings+nudges, in-app report reading view + optional email PDF, voting board, ideas, push (new-audio-ready), WhatsApp group. TESTERS: to install any iOS app a person needs Apple ID + free TestFlight + our invite (Apple's install gate, separate from our magic-link content login). Invite Sol+Colin as EXTERNAL email testers (not App Store Connect users) per Jeff. Beta target ~July 1, ideally July 2-3 (driving on the 1st); build the full scale-ready backend FIRST (long pole). Bundle IDs APPROVED: ai.affirmology.atlas, ai.affirmology.app ("ai" first is just reverse-domain convention, no meaning). ROUND 2 LOCKED (2026-06-24): React Native CONFIRMED for the consumer app (validated by Instagram/Shopify/Discord/Coinbase/Tesla in prod; covers Jeff's "mega cool dynamic" bar; web-reuse is a non-factor for the consumer app); build the LEAN v1 first, rest via OTA fast-follows; FEEDBACK = ratings PLUS written notes on everything; CHATBOT support kept OUT of the oracle chat (separate "Support Oracle" or context-routing); ANDROID dropped from scope (add later only if a tester needs it); Atlas internal-tester route OK but Jeff will elevate several beta users into the Studio as audio-designers (~1 wk out), so design Atlas access to scale past 3; Atlas LOGIN leaning magic-link for BOTH apps (eases creator elevation, drops Cloudflare Access) but NOT final. Atlas wrap is reversible (Jeff may "unwrap" to native later; engine is server-side so it's a face-swap). Honest timeline: lean RN app ~1 week build; 7/2-3 possible-but-tight, safe ~7/5-7 incl. one ~24h Apple beta review; backend is the swing factor, keep it lean. Full analysis: CLAUDE OUTPUTS/Affirmology/Affirmology_AppArchitecture_Decision_v1.md (section 10 = Round 2; supersedes the Capacitor framing in Affirmology_AppDelivery_Plan_v1.md). Consumer spec: CLAUDE OUTPUTS/Affirmology/Affirmology_DeliveryApp_Spec_v1.md. GREENLIT 2026-06-24: Atlas LOGIN = MAGIC-LINK (final, Jeff). Backend start greenlit. BUILD BRIEFS WRITTEN: CLAUDE OUTPUTS/Affirmology/Affirmology_AppBuild_Briefs_v1.md = a paste-ready overnight Claude-Code package: a KICKOFF prompt + Brief A (lean Postgres system-of-record + magic-link auth, extending the existing affirmology-studio API; reuse /api/hermes, /api/library, /api/feedback), Brief B (Atlas Capacitor wrap + magic-link to TestFlight internal), Brief C (Affirmology consumer app in Expo/React Native, lean v1: magic-link login, data-driven library, react-native-track-player offline+background player, scoped oracle chat w/ avatars, feedback+notes+unrated-nudge, in-app report reading view, push). Guardrails baked in (don't touch locked demo, no em dashes, every audio chart-driven, verify w/ pytest + RN build, don't block on Apple/secrets -> log to MORNING-FOR-JEFF.md). Oracle face photos exist per Jeff but NOT in repo -> Code to ask for the folder, placeholders until then. Morning human-only steps: App Store Connect app records, signing/EAS API key, testers, APNs push key, Render Postgres DATABASE_URL if not auto-provisioned, oracle photos folder, relax Cloudflare Access on studio domain. NEXT: Jeff pastes the KICKOFF block into Claude Code on the Mac to run overnight; review OVERNIGHT-REPORT.md + MORNING-FOR-JEFF.md in the AM.
• OVERNIGHT BUILD DONE (2026-06-25): all 3 briefs built, tested, committed; nothing touched the locked demo; no em dashes. Brief A backend (consumer system-of-record + magic-link auth + guardrailed oracle chat, SQLAlchemy Postgres-prod/SQLite-fallback, 12 tests). Brief C Affirmology RN app (new repo affirmology-app/, Expo SDK 56 + TS: login/home/library/offline+background track-player player/oracle chat with the 17 council faces/feedback+notes/in-app report reader; typecheck clean, expo-doctor 21/21, iOS bundle export OK). Brief B Atlas Capacitor wrap (magic-link gate OFF by default so live Studio unchanged; 7 tests). All 17 oracle busts downloaded to affirmology-app/assets/oracles/ (Vedic = Sri Agastya). Reports: CLAUDE OUTPUTS/Affirmology/OVERNIGHT-REPORT.md + MORNING-FOR-JEFF.md. BLOCKERS NEEDING JEFF: (1) install full Xcode + CocoaPods (this Mac has only Command Line Tools) to launch in Simulator -> TestFlight; (2) create Render Postgres + set DATABASE_URL (SQLite works until then, zero code change; Render team tea-d8pf325ckfvc73a7mtqg, svc srv-d8pkp20js32c738r8ofg); (3) Apple: 2 app records, App Store Connect API key for EAS, testers, APNs key. Heads-up: 4 real "enter code" emails hit Jeff during tests (own single-use codes); kill switch shipped (c765e99) so tests can't send real mail.
• ATLAS GOES NATIVE TOO (2026-06-25, Jeff overrode the wrap rec; objection withdrawn): rebuild Atlas in REACT NATIVE (same stack as the consumer app), NOT the Capacitor wrap, because Sol barely uses it + Jeff wants a UI overhaul, so little to preserve. Capacitor wrap (Brief B) PARKED; magic-link auth + gate carry over; backend/engine untouched. CENTERPIECE: the audio-making experience moves INTO the Hermes chatbot and must be STRONG (talk to Hermes -> council composes -> script shown -> iterate by talking -> "render it"; rails already exist via /api/hermes + /api/compose + /api/audio). PRINCIPLE: chat-PRIMARY not chat-ONLY (exact controls reachable in a panel for precision). SEQUENCING: consumer app stays the July priority; native Atlas is a fast-follow reusing the consumer app's RN design system. RUNS ENTIRELY ON EXISTING CLOUD (Render + R2 + Anthropic/ElevenLabs/Fish APIs); NO Mac mini needed (Jeff can't set up the $800 mini yet; render already cloud-side per section 19). Tahoe: Jeff restarts the Mac tonight to bring up macOS Tahoe + install Xcode; verify venv/ffmpeg/node/CocoaPods/Xcode-Tahoe compat after the OS jump. Brainstorm for the Hermes creation experience: CLAUDE OUTPUTS/Affirmology/Affirmology_HermesCreation_Brainstorm_v1.md. Decision detail: architecture doc section 11.
• APP EXPERIENCE NORTH STAR + VISION (2026-06-25, Jeff, big download captured): NEW North Star = keep the user on the app EVERY DAY for the first week. Core = a DAILY SUITE (bedtime, morning, visioning/magnify-day, midday gym/motivational identity track, + a couple of ~10min readings) that forms the week-1 repetition loop. Guiding principle: NAIL IDENTITY first by INTEGRATING all modalities (do NOT silo HD/Gene Keys/astrology/Vedic), then let users BRANCH deep. Journey model = the CONSTELLATION (daily suite lights stars/unlocks; branch into deep journeys that weave readings + affirmational audio + EFT + visualization; a map of lit/unlocked/unexplored; new constellations). Organize user-facing by EMOTIONAL IMPACT + TIME OF DAY + three-step stage, NOT by modality (modality is the under-the-hood source). ADAPTIVE: repeat what lands, replace what doesn't, hit different angles, meet them at their stage + pertinent issue (wealth/relationships/career), informed by chart + chatbot data + a few questions + progress. Custom content cadence driven by monthly/celestial triggers + user-stated "right now" needs, gated by credit cost until cheaper Fish voices land. NAMING: Soul Songs = all affirmational audios; Origin Soul Song = first/demo; The Mirror = the first reading (others may be named individually). ORACLES as educators: personality comes through CHAT, NOT a unique Fish voice per oracle (use the handful of known-good audio voices for now). BUILD LIST locked: all 3 Gene Keys sequences (Activation [anchor first], Venus, Pearl) each as affirmational audio(s) + a reading; full Gene Keys journey later; themed visioning. ATLAS needs per-audio METRICS (identity 1-10, transformation 1-10, energy/intensity, pacing, emotional tone, psych/somatic tools, three-step stage, time-of-day, theme, duration, modality lean, voice/bed, repetition role) usable as creation controls + data + adaptive-engine inputs; tweak directly or via Hermes chat; built for in-car Jeff+Sol creation on the trip. This RESOLVES the palette open questions (oracles teach via chat; build all 3 GK sequences, Activation first; visioning themed; The Mirror = first reading). Full vision: CLAUDE OUTPUTS/Affirmology/Affirmology_AppExperience_NorthStar_Vision.md. First build brief ready: CLAUDE OUTPUTS/Affirmology/Affirmology_Build_Phase1_Foundation_Brief.md.
• VISION ADDITIONS (2026-06-25, Jeff "perfectly nailed it"): FIRST EXPERIENCE = a charted INITIAL PROGRAM (demo -> read their progression + a few intake questions -> initial audios that hit identity + pertinent themes + immediate breakthroughs SO hard they get hooked on usefulness + many-times-a-day listening; same front door even for someone who came "just for Gene Keys"). THE GUIDE AGENT IS NAMED: "THE MINDSET NINJA," one of the founders, her FACE appears throughout (not always, but attributed to her), fun/beloved, weaves in journaling prompts + practices ("take a few breaths, repeat this mantra") + tips + education tastefully; real backend build item (need her face/persona + voice/tone). SYNTHESIS GENIUS = cross-system asides throughout any track ("outside your focus modality, but your Sun in X contextualizes this"); turning the complicated map into usable synthesized info is the core genius. HUMAN DESIGN gets FULL treatment too (a reading + affirmational audios, like the 3 GK sequences): dissect into their HD STORY + usable energy tools; keep deep esoteric (diet/environment/variable arrows) optional/later. CHATBOT = the learning loop that shapes next-day/week audios + custom drops (feel special, retention). BETA TIMELINE = ~JULY 4th; Jeff+Sol refine the initial audios next week + partly on the ROAD TRIP, then version off tester feedback; theme constant ("holy shit this helped me"). UPSELL/GATEKEEPING beyond $20/mo into pricier uses = open workstream (membership doc). INVESTOR: vision translated to pitch language in CLAUDE OUTPUTS/Affirmology/Affirmology_InvestorVideo_AppVisionInsert_v1.md (fold into the investor video brief/master).
• RELATIONSHIPS = core category + VIRAL GROWTH LOOP (2026-06-25, Jeff): users enter birth data for family/lovers/prospective dates/friends/business partners/coworkers; get romantic/business/family/friendship synergy, challenges, clarity in ALL formats (affirmational audios, readings, visual reports, guidance). Engine EXISTS (synastry relationship.py + oracles Eros/Concordia/Hestia/Gaia). VIRAL MECHANIC: a user can't SEND a relationship's revelations to the other person unless that person JOINS; one-button invite (marketing/pitch/opportunity); on joining, that relationship piece is ALREADY UNLOCKED so the new member lands in a personalized "holy shit" moment about a real person in their life = built-in high-intent compounding referral. Consent/privacy: entering another's birth data is for the user's own exploration; SHARING requires the other to join (the consent gate); keep invites tasteful. Captured in vision doc section 7b + investor insert.
• DELIVERY CONVERGENCE (2026-06-24, 84843c5) [Jeff's failsafe rule: multiple delivery methods so audio from too many sources is never lost]: every audio, from ANY source (a Cowork session, either Claude Code window, Hermes, a push from Sol/Colin), MUST be registered through ONE door so it fans out to BOTH email (at-computer) AND the Listening Room / Atlas (out). THE DOOR: POST https://affirmology-studio-api.onrender.com/api/library/ingest, Bearer = cloud STUDIO_AUTH_TOKEN, body {person_id, title, source, context, mp3_url (or mp3_base64), pdf_url (or pdf_base64), email:true}. It stores a 'done' Listening Room entry AND emails the recipient. RULE FOR EVERY CHAT: if you produce an audio, call this endpoint; do NOT paste a raw link (raw links make a chat-embedded player that fails for Jeff). Pipeline renders already auto-email + land in the Room; this door is for everything born outside the pipeline. The Listening Room itself is the Studio "Listening Room" page (GET /api/library) with inline play + Love/Needs-work/Approve. See memory [[always-email-audio-to-jeff]]. Fish voice tests were backfilled through this door.
• FISH VOICE + SPOKEN READINGS/AFFIRMATIONS (2026-06-24, ad-hoc via /tmp/voice_reading.py, NOT yet a committed engine feature, build it in later): Fish Audio TTS is wired (FISH_API_KEY in affirmology-agent/.env) to voice council readings + affirmation tracks cheaply. Voices LOCKED: Storyteller reference_id e686ae649ee44f219a108aacba206c1a for READINGS (Jeff's pick), Emily a325095a7cc049cebf39b1de9464fc73 (meditation) for DAYTIME/SLEEP; Fish model 's1' + prosody speed 0.85-0.92, sentence-level pauses. PROVEN "reading to audio" FLOW: a long all-systems reading is generated as 7-8 council sections (POST /api/hermes per section, as_person jeff/sol; CLEAR THE THREAD between each via /api/hermes/clear or it throws BadRequestError; sections western, asteroids, vedic, humandesign, genekeys, purpose, threads, timing) then stitched (STRIP oracle meta intros like "Sophia: came in across two calls") then Fish-voiced then delivered. A full all-systems Master Reading runs ~70 min, so position it as the PREMIUM/UPSELL tier (entry ~10 min focused, mid = single-system). Daytime affirmation tracks = first-person script, Fish-voiced, looped 3x, mixed UNDER the Heavenly Circuit bed (affirmology-agent/music/Heavenly Circuit.mp3). Optional sound treatment (whisper pad + soft bowl at the section turns) exists but underwhelmed Jeff; a real continuous bed will likely come from Suno; a bed-options demo reel is parked. DELIVERED 2026-06-24: Jeff's full Master Wealth Reading (clean + treated), and Sol's two SURPRISES to solballard@gmail.com (her Master Wealth Reading + her "I Am Already This" Daytime Anchor with the Heavenly Circuit bed). All in the Listening Room.
• STREAMING PLAYER FIXED (2026-06-24): Safari could not STREAM media.affirmology.ai audio (download + email attachment worked, only streaming failed) because the affirmology.ai zone had BROWSER INTEGRITY CHECK on (Bot Fight Mode was a red herring; it is dashboard-only on this plan and not the cause). FIX applied via the Cloudflare API: set zone browser_check=off + purge_everything. Jeff confirmed Safari streaming now works. So delivery is back to PLAYER-LINK as default: media.affirmology.ai/player.html?src=URL&title=...&ctx=... (a public branded HTML5 player; web/player.html is mirrored to R2 key player.html), attachments only as a fallback. No more ~25MB attachment ceiling; the Listening Room / Atlas now streams in Safari too.
• KEYS: Claude holds in gitignored affirmology-studio/.env: RENDER_API_KEY, a broadened CLOUDFLARE_API_TOKEN (Zone Settings + Cache Purge + DNS + Workers + R2; still NO Bot Management, which is dashboard-only on this plan), RESEND_API_KEY, FISH_API_KEY (also in affirmology-agent/.env), plus cached CLOUD_STUDIO_AUTH_TOKEN + RENDER_SERVICE_ID. Per Jeff's 2026-06-24 request to hold keys and minimize his manual dashboard work.
• LAST APPROVED: 2026-06-23, Jeff approved the C26 pacing (slow DEEP ~0.88, ACTIVATION ~1.03 "for daytime"; memory c26-deep-pacing-approved) and greenlit C3.
• FUNDING WORKSPACE STARTED (2026-06-25, Cowork, NO code): new internal working doc Affirmology_FundingStrategy_Memo_v1.md captures Jeff's evolving funding thinking cross-referenced against existing TermSheet_v1 / BuildVsHire_UseOfFunds / CashCowReturns / MoneyList / RoundContingency. KEY DELTAS vs on-paper: (a) Jeff now leaning a leaner staged Round 1 ~$75-150K (possibly two tranches, favorable first / higher-valuation second) instead of the documented single $150-250K @ $1.5M post; valuation math unchanged ($15K=1%, $5K=0.33%). (b) Convertible-note question RESEARCHED + marked POSSIBILITY only: a SAFE does not fit an LLC and convertible notes in an LLC carry COD-income + basis-drop tax traps (counsel generally says avoid in an LLC or convert to C-corp first); clean LLC options = membership units / straight promissory note / profits interest; recommendation = let each investor choose equity, loan, or both (already in the term sheet) rather than a hybrid; only do a true convert-or-repay note as a counsel-papered one-off if an investor insists. (c) Debt-vs-equity: given Jeff's cash fragility, lean equity cushion over a debt repayment cliff during the launch window. Sol up-to-25% + Colin up-to-5% advisor+finder already match existing earn-in/finder structures; future-CEO (Jackie 10-25%) is a real later dilution event to model. ELEVENLABS STARTUP GRANT researched + full application drafted (Affirmology_ElevenLabsGrant_Application_v1.md): 33M chars / ~680 hrs voice AI free for 12 mo, under-25-employees product-with-monetization bar = strong fit; it is CREDITS not cash (offsets the voice-render line + partner relationship, does NOT replace the raise); apply at elevenlabs.io/startup-grants, ~1 wk decision. NOT going to Colin yet (Jeff: prep more first, then send for review). OPEN for Jeff: confirm Round 1 number/tranche shape; define Sol 20->25% trigger; confirm Colin base 2-3% + finder; park or kill the convertible note; price the app-ops line + Korea living + demo-kit; submit the ElevenLabs app.

• LAUNCH CRM PLAYBOOK BUILT (2026-06-26, Cowork, NO code): new Affirmology_LaunchCRM_Playbook_v1.html is the operating plan for the Jun 29 warm outreach wave + the Airtable CRM. Reconciles the existing launch docs (DemoLaunchPlan v2, LaunchRunway 12-Day, NDAGate Design, MutualNDA, LaunchTimeline, TimingAndLaunch ActionPlan) into ONE funnel: 1:1 WhatsApp invite -> demo behind the existing Mutual-NDA e-sign gate -> same vision video -> self-sort (Tester/Investor/Connector as independent flags) -> 1:1 convo -> Jul 4 free beta -> escalate top testers to the Studio "Creator/Community" tier. KEY DECISIONS IN THE DOC: (a) Airtable is the SYSTEM OF RECORD (GHL demoted to send-only or retired, Jeff to confirm); (b) Jun 29 send is WARM-only (consistent w/ Mercury Rx Jun 29-Jul 23 favoring warm beta invites); COLD investor outreach holds until after Jul 23; Jul 4 beta strictly FREE/gift-framed (no paid/founding upsell that day per the locked timeline). INCENTIVE STRUCTURE designed (Jeff's 2026-06-26 download) + TIERS NAMED (Jeff, 2026-06-26): Floor (List Member, couple months free, always welcome to buy) -> INNER ORBIT (the first beta testers: free thru beta + founding status; earned usage milestones add free time stacking to lifetime-at-a-tier + affiliate + custom Soul Songs; owes usage+feedback) -> THE CONSTELLATION (the creator membership tier: design audios in the Studio WITH us and FOR us via chatbot + controls + their own voice notes; make their OWN audio journeys + audios for their communities; affiliate/rev-share; the charter cohort upgraded from Inner Orbit = the FOUNDING CONSTELLATION; also open to outside creators later; live in-app chatbot audio-creation is a fast-follow but the Constellation distinction holds because they make+publish, not just listen). KICK-OUT = soft downgrade Active->Warning->Downgraded-to-Floor, Jeff approves. COMPLIANCE FLAG: cash affiliate for referred USERS, but investor-intro rewards stay NON-CASH (finder/broker-dealer risk; route finder deals through Colin/counsel). AI CONVERSATION TRACKING: no WhatsApp/Airtable MCP in the registry yet, so the flow is WhatsApp Export chat/forward/screenshot -> paste into Cowork -> AI parses person+summary+intent+sentiment+next-step+named-connections -> writes Interaction + updates People (directly once an Airtable MCP is added via a personal access token, else import-ready rows). Demo-taker cross-ref = match engine/Listening-Room emails to People, flip Stage=Took demo. Airtable schema = 5 tables (People spine, Interactions AI-log, Beta Testers, Investor Pipeline, Audios) delivered as 4 import-ready CSV templates (Affirmology_Airtable_{People,Interactions,BetaTesters,InvestorPipeline}_v1.csv). Doc ends with a categorized checklist + 6 open decisions for Jeff (free-usage numbers, lifetime tier scope, affiliate cash, investor-intro reward, GHL fate, who owns the CRM daily). CAPITAL MODEL CLARIFIED (Jeff, 2026-06-26, hard correction): there is NO cold raise, NO public round, EVER. ALL capital is WARM, from people Jeff knows or one warm step into his network, inside conversations already underway. Jeff WILL and MUST take money BEFORE Aug 11 (needs cash now); taking a warm check during the Mercury Rx is fine (the Rx only cautioned against cold-INITIATING strangers, which he is not doing). Aug 11-13 is reframed as the formal CLOSING / membership-unit ISSUANCE date (money comes in before; units issue at closing, riding the launch), NOT "when the raise opens." A beta tester who connects him to someone = a warm path he keeps working. Likely instrument (FOR COUNSEL, not decided): short subscription/investment agreement signed now (amount, units or loan or both per investor-chooses, units issue at the Aug closing), funds in now, member ledger updated at closing; gating facts = LLC must be filed (Jun 26/28) + operating agreement must DEFINE the units first; NO convertible notes (Jeff, hard rule). LLC FILED 2026-06-25 effective 2026-06-26 (done), bank account Mon/Tue so cash can be taken straight up. THE ROUND STAYS OPEN / ROLLING, it does NOT close: capital arrives on each investor's own timing across the warm phase, and units are issued in BATCHES at EXACTLY TWO "ISSUANCE CEREMONIES" (Jeff confirmed 2026-06-26, both dates, NO future dates): Aug 11-13 launch + Sep 11 solar return (multi-birthday principle applied to capital); whoever funded by each date is issued in that batch, money in flight between them lands in the Sep ceremony. NO premature close but BOUNDED (not open forever). This is BRIDGE capital to fuel through the end-of-Sept launch + the scale it takes; by Sept, ads + event customer/affiliate enrollments cross $20k/mo and climbing, so REVENUE takes over and the raise winds down. OPERATING AGREEMENT is an INTERNAL doc (adopt/execute this weekend with Sol, NOT filed with the state). SEQUENCE: vision/intent (the Jun 28 mission video + high-level story, and an ~80pg company report at the vision level) can go out BEFORE the OA exists (intent landing is what matters there); but specific equity language (percentages, "this many units"), the detailed terms report quoting those numbers, any SIGNED investment doc, and MONEY changing hands wait until the OA defining units is in place. Likely instrument = a short subscription/investment agreement (units or loan or both per investor-chooses, units issued at the next ceremony); a bespoke instrument only if a specific investor wants one, counsel-papered then. Agreements counsel-papered even among friends (Colin owns). Doc updated to remove all cold/strangers/angel-raise/convertible-note framing and to add the issuance-ceremony (rolling-round) concept (sections 7.2/7.3 + timing table Sep 11 row + Investor table "Units issued at" field). RESOLVED (Jeff): use BOTH dates as two ceremonies, no electional pick needed. ALSO: GHL is RETIRING (Airtable is the sole record). SEND DATE moved to SUN JUN 28 (not Mon 29) per a fresh Hermes/Sophia electional reading (2026-06-26, funnel healthy): publishing finished materials to a WARM list at the station is resonant (Mercury stations ~3deg Cancer on his natal Moon = homecoming not solicitation); 28 beats 29 because Mercury still direct on the 28th (forward momentum when it lands) + Jupiter enters Leo Jun 30 (be already in conversation when it fires) + Sunday-evening beats Monday inbox; caveat = Neptune square Moon can blur the ask, so sharpen the terms / read aloud first. AIRTABLE MCP path confirmed (official hosted MCP at https://mcp.airtable.com/mcp, add as a custom connector in Claude desktop + OAuth or a scoped personal access token); no WhatsApp personal MCP exists so conversation logging stays Export-chat -> paste. ASSET FLOW REFINED (Jeff, 2026-06-26): NOT a single forced path. The one pasted link opens a front door with TWO clear doors = "Start with the experience" (demo, primary) and a quiet "Watch the 2-min vision first" (for investor-minded who won't fall for the demo cold), with a deeper link from the vision to the company report. After the demo audio, a SHORT ~1-min HYPE VIDEO plays. Every person gets a PERMANENT PERSONAL HUB (their own tokenized link) holding their audio + the hype video + a "learn more about the mission" button + a "talk to Jeff / become founding member" CTA, delivered by BOTH email and text so nothing is lost and they can return/forward; follow-ups update the same hub. HUB IS TECHNICALLY ONE TEMPLATE PAGE, NOT A PAGE PER PERSON (Jeff asked): same pattern as the existing player.html (single static page personalized by URL); a hub.html on the demo site reads a per-person TOKEN in the URL (e.g. demo.affirmology.ai/h/) and JS fills in that person's audio (already a permanent R2 link), name, vision video, doc links, and a "send the demo to a friend" share button (built-in referral/viral loop); everyone goes to the same demo site, the token is the only difference, and that token is also what powers the seen/not-seen tracking. NOT built yet. SEEN/NOT-SEEN TRACKING added: tokenized links + Resend open/click + engine listen events feed People fields (Link clicked, Door taken, Demo listened, Vision watched, Report viewed, Last seen, Hub link) driving a NUDGE LADDER (each stall has a specific non-naggy nudge; the demo nudge becomes a promise = "something incredible after you go through it"). Playbook sections 3.1/3.4 + 9.4 + People table + checklist C/G updated. HARD RULE (Jeff, 2026-06-26, firm "do not do this"): NEVER ask anyone about investing (or any intent/routing question) at the gate or before they have EXPERIENCED the demo + mission. They have felt nothing at the gate, so asking "are you here to invest" is backwards and kills it. The gate captures ONLY name + email + NDA signature, nothing about intent. Self-sort EMERGES AFTER the experience: primarily from their reaction/conversation (AI-logged), secondarily from a SOFT post-experience prompt on the personal hub. "Self-sorted as" is never populated from a gate question. Playbook section 5 retitled "Self-sort: after the experience, never before" with a stop-callout. NEXT: Jeff answers the 6 decisions; get clear with counsel on the warm-capital instrument BEFORE it goes in any contract; finalize the OA with Sol this weekend; connect Airtable MCP then Claude builds the base; counsel pass on NDA + draft Beta Tester Agreement; build the two-door front door + personal hub + 1-min hype video.

• RESOURCE HUB BUILD BRIEF WRITTEN (2026-06-26, Cowork, NO code): Affirmology_ResourceHub_BuildBrief_v1.md is a paste-ready kickoff for ANOTHER chat (Claude Code on the Mac) to build an internal, GATED HTML resource hub at a NEW subdomain resources.affirmology.ai (Cloudflare Pages + Cloudflare Access allowlist jeff@jeffparker.love + Sol solballard@gmail.com + Colin cjacobsandassoc@gmail.com) for the team to read the project's docs as clean navigable HTML. House-style reference = Affirmology_LaunchCRM_Playbook_v1.html (extract its CSS). A CRON (launchd, every 2h on the Mac) runs a Python generator that hash-detects new/changed docs, Haiku-extracts title/summary/category/related per doc into a manifest.json, regenerates index + category + per-doc pages, and wrangler pages deploys. Decoupled from the studio on purpose (studio is being retired). Doubles as the doc reorg. Jeff is handing the brief to another chat to execute. (hub.html per-person launch landing page = PARKED, Jeff lukewarm.)

• NDA GATE v2 BUILT + DEMO-ENDING SPEC (2026-06-26, Cowork): (1) affirmology-site/nda-gate-v2.html upgrades the existing in-page demo gate into a proper confidentiality experience: captures NAME + EMAIL (email = CRM identity key; old gate took only name), shows a warm summary + expandable FULL "Demo Preview NDA" terms (3yr, no screenshot/record/share, no license, FL/Miami-Dade, counsel-review draft based on the Mutual NDA + the NDAGate design draft), typed-name signature + agree checkbox (clickwrap = name+email+timestamp+UA, defensible), records to POST /api/nda with localStorage fallback, fires funnel tracking (gate_viewed, nda_signed via /api/track with the ?id= token), carries name+email into the chart form, NO intent/routing questions (hard rule). Reuses the live index.html brand tokens, integration-commented. KEY ARCHITECTURE CALL (recommended to Jeff): the DEMO gate is an IN-PAGE CLICKWRAP and does NOT need an external signable-doc website / DocuSign template; it records itself to the CRM. Reserve the external e-sign tool (DocuSign PowerForm / Dropbox Sign + template + webhook) for the higher-stakes BETA TESTER AGREEMENT + INVESTOR docs, NOT the demo gate (a heavy e-sign flow before the wow would add friction and kill conversion). To make the gate actually write to the CRM, the demo Worker needs POST /api/nda + /api/track routes that write to Airtable (Airtable token as a Worker secret). (2) Affirmology_DemoEnding_Spec_v1.html = the ending/offboarding + accumulating personal page spec: after the song + hype + mission/vision video, FOUR equal ending paths (Beta "I want to use this" -> short beta-info page; Investor "tell me more" -> full company report/data room; Connector "I know someone" -> light capture; Keep-warm "keep me posted" -> Floor nurture), each tracked. The PERSONAL PAGE accumulates: Day 0 song + videos + 4 paths + share; Day 1 Blueprint Reading PDF + MP3 reading; Days 2-5 optional extra custom audios (morning + daily activation); optional Sol+Jeff founders' video. Drive hard to the PDF. Advice captured: stage the extra audios (don't dump), gate extras to engaged users (cost), keep experience-first before the written investor deep-dive, one permanent tokenized page (not per-person). Counsel review of the NDA text still needed. NEXT for Jeff: finish the Airtable token/MCP; decide clickwrap-gate (recommended) vs external e-sign at the gate; then a build chat wires /api/nda + /api/track on the Worker and integrates nda-gate-v2 into index.html.

• NDA STRATEGY CHANGED -> FULL E-SIGN FOR EVERYONE (Jeff + Sol's hard demand, 2026-06-26): the demo gate is NO LONGER a lightweight clickwrap. ANYONE who sees the demo must sign the FULL mutual NDA via real e-signature BEFORE entering (more robust for sending to ~100 people: testers, investors, allies). The in-page clickwrap (nda-gate-v2.html) is SUPERSEDED for the gate. Jeff uploaded the approved signed NDA (the one Colin Jacobs already signed, his Jeff Parker LOVE letterhead). Built Affirmology_NDA_Signable_v1.pdf = a clean, blank, REUSABLE version of that exact text (verbatim 13 sections), Party A prefilled (Jeffrey Parker, jeff@jeffparker.love), Party B/recipient fields OPEN (signature/printed name/email/date), Effective Date = "as of the date of signature below" so it works for self-serve signing. This PDF is what Jeff loads into the e-sign tool to make a reusable self-serve signing link. NEXT (e-sign setup, hand-hold like Airtable): pick the tool (DocuSign PowerForm = fastest/recognized, or Dropbox Sign = embeddable/branded; Bloom.io was the old manual tool), upload the PDF, place the recipient name/email/signature/date fields, generate ONE reusable signing link / PowerForm, wire its webhook -> Airtable (NDA signed + name/email), and gate demo access on completion (token redirect or Cloudflare Access allowlist). The branded "review & sign to enter" gate page (replacing the clickwrap gate) is still to build (task open). Also offered: editable DOCX for counsel.

• E-SIGN TOOL = SIGNWELL (Jeff, 2026-06-26, NOT DocuSign). NDA MADE MORE ROBUST: Affirmology_NDA_Signable_v2.pdf (4pp, 20 sections) supersedes v1, adds: No Recording/Reproduction (no screenshots/screen-record), Non-Circumvention (2yr, protects investor/partner/customer relationships), No Reverse Engineering/No Competing Development, Feedback License (perpetual royalty-free to Affirmology), Representatives definition + liability, attorneys' fees to prevailing party, Assignment/Successors (Jeff may assign to a successor on incorporation/financing), Notices, Waiver/No-Partnership; Party A prefilled, recipient fields open, effective on signature. Still a counsel-review draft. SignWell flow: upload the v2 PDF as a Template, place recipient Signature/Name/Email/Date fields, generate a reusable TEMPLATE LINK (bulk self-serve signing), API key via X-Api-Key header (Settings -> API), webhook "document completed" -> Airtable (NDA signed) + grant demo access. Hand-hold guide = Affirmology_SignWellSetup_Steps_v1.html. WAITING ON JEFF: paste the SignWell Template Link + API key (Claude will store the key like the others + wire it).

• AIRTABLE STATUS (2026-06-26): Jeff created the base + a PAT + the MCP connector. PAT stored by Claude in gitignored affirmology-studio/.env as AIRTABLE_API_KEY. BLOCKER: the PAT authenticates (HTTP 200) but returns {"bases":[]} = the token has NO base access yet, so Claude cannot build the CRM into it. FIX for Jeff: go to airtable.com/create/tokens, click the "Claude CRM" token, under Access add the base "Affirmology Launch CRM" (or add the whole workspace), save; then Claude retries. (No Airtable MCP tools surfaced to the Cowork session yet either; the PAT-over-API path is the working route once base access is granted.)

• POST-DEMO STRATEGY REFINED (Jeff, 2026-06-26, logged into Affirmology_DemoEnding_Spec_v1.html): (1) INVESTOR DOC SHOWN INLINE, not behind a click: the company write-up is good enough to display right on the page under the video for everyone, because clicking "see the investor report" feels like an affirmative step toward investing and intimidates; in plain view (zero commitment) more people read it, and reading moves them. The investor PATH becomes a soft "open to talking," not "go see the report." (2) NEVER-GATEKEEP RULE (hard): once a person is in, their personal page is theirs FOREVER, NO new NDA, NO login wall, NO lockout, ever; they signed once at the door, from there it is frictionless; accept the small share risk (nobody hands their soul page to a thief; if they show a friend it is excitement = free marketing); never lock them out of their own info. (3) Morning-after = their Blueprint Reading PDF + MP3 reading auto-added to the page. (4) DAYS 2-5 EXTRA AUDIOS (logged idea): email engaged people (those who gave their email) short extra audios on different use cases over the next few days (morning activation, daily activational, wind-down), each added to their page, to keep it fresh and reward engagement. (5) BETA INFO PAGE drafted (the "I want to use this" destination): tester-facing copy capturing Jeff's expectations (actually test + give feedback; full app features unlock in coming weeks; the start is testing audios in daily life; listen several times a day, nightly at bed, on waking, a few through the day, test versions, give feedback; join the WhatsApp chat) + what they get (early/founding access, growing audios, a real voice, free month or two at launch just for joining) + the honest "if inactive we nudge then may free your spot, still love you, you keep the launch months" + the incentives-soon (create your own audios in the chatbot, a creator account, make for your community, free tiers = lifetime for diving in or at least a year). Full incentive framework still lives in the Playbook section 6 (Inner Orbit/Constellation); the beta page just shows the warm simple promise. (6) BETA OPT-IN = TWO BUTTONS (Jeff, 2026-06-26): "I am in" = one click that puts them in the system WITH the commitment (a plain-language "I will use it + give feedback" click, NOT a legal doc) -> committed Inner Orbit record (new CRM field Beta committed + date); the formal BETA TESTER AGREEMENT is sent to sign LATER via SignWell (when beta opens), so a second signature never interrupts the excited moment (they already signed the NDA at the door). Second button "I would like to talk first" = opens a conversation with Jeff / "let me know" ping, no commitment. Principle: commitment now (while hot), paperwork later.

• AIRTABLE CRM BUILT (2026-06-26, Cowork, via the Airtable MCP). Jeff's manually-created base was invisible to the token (empty bases), but the MCP could see his workspace (wspAjqGbLClKnjMUR, "My First Workspace", owner), so Claude CREATED the base directly: base id appSuTUcOQbpkxLUR "Affirmology Launch CRM" with all 5 tables + linked records: People (tblOOWgEjVfX21NT3, master, ~26 fields incl funnel/tracking), Interactions (tblvMUdNEZ6s5YHHR, AI log, linked to People), Beta Testers (tbljp092kPCME6G0o, tiers Floor/Inner Orbit/Constellation + usage status + Beta committed, linked), Investor Pipeline (tblH250hyGVM3dpPN, stages incl Agreed/funded + Units issued, Funds received, Units issued at Aug/Sep, Person + Referred by links), Audios Delivered (tblkYYF0LucPyLK7Q, linked). Jeff can DELETE the empty base he made by hand (the real one is the Claude-built "Affirmology Launch CRM"). NEXT on CRM: load Jeff's people list; wire the NDA/demo/Resend feeds; build filtered views (custom views need the interface tools, default Grid views exist).

• AFFIRMOLOGY LLC EXISTS NOW (Jeff, 2026-06-26): the LLC is formed (the Jun 26/28 filing happened). So the NDA's Party A is now the COMPANY, not Jeff individually. NDA regenerated to Affirmology_NDA_Signable_v4.pdf: Party A = "Affirmology LLC, a limited liability company, by Jeffrey Parker, its authorized member" (conformed /s/ signature pre-filled so Jeff signs ONCE for all, never per-recipient); recipient block SIMPLIFIED to just Signature + Date (the printed-name + email lines removed) because SignWell auto-captures the signer's name, email, IP, and timestamp in its completion certificate (exactly as the Colin-signed copy showed). v4 supersedes v1/v2/v3. (Confirm the LLC's state of formation, presumably Wyoming per the launch plan, for the final; governing law stays Florida/Miami-Dade per the doc.) NOW v5: Affirmology_NDA_Signable_v5.pdf embeds Jeff's REAL signature image (Signature Jeff.jpg in the folder) on the Party A line above "By: Jeffrey Parker, Authorized Member"; recipient block = Signature + Date only. v5 is the doc to upload to SignWell. SignWell field setup confirmed with Jeff: top = a "Name" field type for Party B (identifies them, auto-fills from the signer's entry on the template link), bottom = Signature + "Date Signed" (auto). AUTOMATION BRIEF for Claude Code written: Affirmology_NDAFlow_BuildBrief_v1.md = paste-ready kickoff to automate the whole pipeline (SignWell template + reusable link + webhook -> upsert Airtable People by email, set NDA signed/at + Stage, grant demo access; plus /api/track funnel events; demo-taker reconciliation). Includes the live Airtable IDs (base appSuTUcOQbpkxLUR, People tblOOWgEjVfX21NT3 + field IDs) and the PREREQUISITE that Jeff's PAT must be given access to the new base (add the base to the token at airtable.com/create/tokens; it was created via the MCP so the old token shows empty). WAITING ON JEFF: paste the SignWell API key + Template Link; add the base to the PAT.

• WEBSITE OWNERSHIP (2026-06-26): THIS Cowork chat is now responsible for the MAIN WEBSITE (affirmology-site/, the demo: on-screen sequence -> intake form -> NDA gate -> generation screen -> result). A SEPARATE chat builds the investor video + the other site. NEW SEQUENCE: a "Joe and Monica" video plays DURING audio generation (goes on the generation/primer screen, affirmology-site/primer.html) - asset still needed from Jeff before wiring. FORM CHANGE DONE: added a required "Sex" field (Female/Male) to the index.html intake form, threaded into the submit payload + localStorage, and added a matching "Sex" singleSelect field to the CRM People table (fldafEcjKcU4JBDU7) so it is tracked. INVESTOR DOC NAMING (open, Jeff deciding): the inline post-demo company document needs a name that does not shy from investment but appeals to non-investors too; Claude recommended "The Affirmology Mission & Opportunity" (on-page link "The Mission, Vision & Opportunity"), alternatives "Mission & Investment Brief" / "Inside Affirmology". NEXT for the website chat: name the doc, get the Joe+Monica video to wire into the generation screen, integrate the SignWell gate per the NDA-flow brief.

• EASING-IN PRIMER SEQUENCE WIRED INTO THE LIVE FLOW (2026-06-26): affirmology-site/primer.html REPLACED with the EasingIn sequence (from Affirmology_PrimerIntro_EasingIn_v1.html); old primer backed up to affirmology-site/primer_legacy.html. Flow: index.html form submit -> primer.html#<jobId> -> black trailer cards ("Your blueprint is being read / Take a breath / While it loads, watch this") -> tap BEGIN (the user gesture that unlocks sound-on autoplay) -> Joe & Monika experience film WITH SOUND -> on end, "Yours is ready" card -> BEGIN -> result.html#<jobId> (the player). Wired: reads jobId from the hash, polls GET /api/status/<jobId> every 4s and calls markReady() when status is done/ready/complete or an mp3_url appears (orrery loop holds on the ready card if generation runs long; film ~3:40 > generation ~2:00 so it is normally ready first), and carries the jobId to result.html. TWO THINGS PENDING: (1) VIDEO HOSTING + EXACT FILE: videoSrc is currently the R2 placeholder https://media.affirmology.ai/experience/affirmology-experience.mp4; the real Joe & Monika / "updated Affirmology Experience Video" (Jeff says he added an updated one to the main folder; the clear candidate is Brand/Affirmology Joe and Monika Video.mp4, 335MB) must be UPLOADED to R2 at that key, and Jeff must confirm the exact updated filename. orrerySrc also points to an R2 placeholder. (2) the SignWell NDA gate still to integrate per the NDA-flow brief. CRM TRACKING for Jeff's "log who I send the demo to" workflow: added People fields "Demo sent" (checkbox fldmJbfzQ38D87PpT) + "Demo sent at" (date fldz9Dv2d3SM4gmvK); pair with the existing Demo listened / Link clicked / Stage funnel fields to track conversion. Jeff will add the names of people he sends the demo to; Claude can load them on request.

• VIDEOS LIVE ON R2 + WIRED (2026-06-26/29). Decision: SELF-HOST on R2 (not YouTube) for confidentiality (NDA-gated) + cinematic control; R2 egress is free. Code on the Mac encoded the films web-light (H.264 + faststart) and uploaded the two big ones. All three keys now stream (HTTP 206, video/mp4) at media.affirmology.ai: experience/experience-video.mp4 (the Joe & Monika experience film, wired into primer.html), experience/mission-video.mp4 (the mission/investor video), experience/orrery-loop.mp4 (the 4MB ready-card fallback, uploaded by Claude from the sandbox). MISSION VIDEO PLACEMENT DECIDED + WIRED (Claude, owns the post-demo logic): added a "Where this is going" section at the END of result.html's #afterlisten onward area (after their song + "that was a taste" + report/downloads) = the mission video, then the Mission & Opportunities doc shown INLINE via iframe (per the no-click rule), a PDF download, and the four soft next-step buttons (path-beta/invest/connect/keep, currently href="#" STUBS to wire next). CAUTION: Claude edited result.html's onward block; coordinate so the Code chat does not also edit that section (collision risk). MISSION & OPPORTUNITIES PDF: Jeff getting it from the other chat; the page references https://media.affirmology.ai/docs/mission-and-opportunities.pdf (inline iframe + download). When Jeff drops the PDF in the main folder named mission-and-opportunities.pdf, Claude can upload it from the sandbox (small) to that key (ContentType application/pdf) and the inline doc lights up. NOTE: site edits (Sex field on index.html, new primer.html sequence, this result.html onward block) are LOCAL only; demo.affirmology.ai deploys from the GitHub repo, so they must be committed + pushed before the LIVE run reflects them.

• PERMANENT PERSONAL PAGE (HUB) + EMAIL DRIP + DIRECT ASKS (Jeff's big download, 2026-06-29). THE VISION, to build out (result.html is becoming this hub): after the mission video, the page is PERMANENT, each person reaches it forever via a key-code/token, no re-gating. On it: (a) rewatch the Joe & Monika experience video; (b) a CLEAR player for their Origin audio + a download button (Jeff is fine with download; would love them to keep returning but download is OK); (c) the Mission & Opportunities as an ENTICING CARD/BUTTON, not the full embedded doc (the doc itself runs too wide) -> DONE on result.html (gold card "Read it"); (d) a Sol+Jeff video (they record tomorrow, a few minutes in their own words about what excites them; upload when ready); (e) the page UPDATES as new audios drop over the next days. DIRECT ASKS on the page (not shy) -> DONE on result.html: Investors (buy-in from as little as $5,000 to fund the mission), Beta testers, and Connections (esp. technology or team). EMAIL DRIP (to build): email gives them the permanent link + "for the next couple days we'll surprise you with new audios coded to you, watch your email." Cadence CAPPED at ~3 MORE emails (Jeff, 2026-06-29): (1) next morning = their Origin Blueprint PDF + The Mirror reading; (2) and (3) one new audio each on the following days; plus room for a surprise BONUS. The page/first email should TELL them what is coming (a few more coded gifts over the next days) so they watch their inbox. Each drop updates the hub (+ link to the report). Purpose: stay on their radar, give value, build excitement. EMAIL FROM/RETURN ADDRESS: must be an account we own, e.g. hello@affirmology.ai (currently FROM=reports@affirmology.ai; affirmology.ai is verified in Resend so any @affirmology.ai sends; ensure Cloudflare Email Routing forwards hello@ to a real inbox so replies land). TRACKING/HEATMAP (to build): record everyone's email opens, page usage, and HOW MUCH OF THE VIDEOS they watch, etc. (tokenized links + Resend open/click events + page + HTML5 video timeupdate progress events -> CRM funnel fields).

• BETA PAGE BUILT (2026-06-29): affirmology-site/affirmologist.html = small on-brand "Become an Affirmologist" page. Copy kept LIGHT per Jeff (no specific incentive numbers yet): be in the first circle, actually use it + give feedback, full app features open over coming weeks, there will be gifts/incentives + a "Founding Constellation" badge that stays on their account forever, details + a simple agreement follow later this week. Signup form (name+email) POSTs to /api/beta-signup (to be wired to Airtable: Is Tester + Beta committed) with graceful thank-you fallback; carries the ?id token; includes Jeff's real Calendly https://calendly.com/jeffparkerlove/speakwithjeff (secondary, since most contact is by text). Jeff reviews/edits the beta copy tomorrow; he + Colin finalize incentives. Linked from result.html "Become an Affirmologist" button.

• SITE COMMITTED, NOT PUSHED (2026-06-29): committed locally as 77663fa on master (primer.html easing-in sequence, result.html onward block, affirmologist.html). The SANDBOX CANNOT PUSH (no GitHub auth here; index.html sex-field change was already committed earlier by the Mac/Code). TO GO LIVE Jeff/Code runs on the Mac: cd /Users/jeffreyparker/CLAUDE/AFFIRMOLOGY/affirmology-site && git push origin master. Mission & Opportunities PDF still pending from the other chat (page links to media.affirmology.ai/docs/mission-and-opportunities.pdf; drop the file in the main folder named mission-and-opportunities.pdf and Claude uploads it as application/pdf).

• MISSION & OPPORTUNITY PDF identified (2026-06-29): it is in Jeff's Google Drive = Affirmology_Mission_and_Opportunity_v1.pdf (5.3MB, formatting still WIP, "use for now"). It must be uploaded to R2 key docs/mission-and-opportunities.pdf (the result.html card + download point there). Path: drop it into the main AFFIRMOLOGY folder renamed mission-and-opportunities.pdf and Claude uploads it (small, seconds), OR the Code chat pulls it from Drive and uploads. Until then the result.html mission card 404s (expected).

• EMAIL BEAUTIFICATION (Jeff, hard rule, 2026-06-29): ALL emails that go out (the demo delivery email + the 3-email drip in Affirmology_DemoDripEmails_v1.md + any others) MUST be sent as BEAUTIFUL BRANDED HTML on the existing emerald/gold template (api/emailer.py), NEVER plain text, from an owned address (hello@affirmology.ai). The drip doc holds the copy; wrap it in the branded template before sending.

• DEMO FLOW FIXES (2026-06-29, from Jeff's live test, NOT yet pushed). (1) ONE TAP after the film: primer.html now sends them to result.html#<jobId>&film so result.html's filmMode triggers; filmMode rewired to SKIP the old tap-to-begin overlay + the old #waiting loading teaser ("vector loop") entirely and go straight to the single "Play my audio" ritual. (2) ORRERY DURING AUDIO: added #orrery-visual (media.affirmology.ai/experience/orrery-loop.mp4, muted/loop) to #delivered; it plays as the visual when the song plays. (3) AFTER THE SONG STAYS IN PLACE: revealAfterListen() no longer navigates to the OLD whatsinstore.html vector reel (that dead-ended at downloads); it now reveals result.html's in-page #afterlisten (outro + report + downloads + the mission video + asks + warm note + Become-an-Affirmologist), and renders the blueprint. whatsinstore.html is OUT of the flow. (4) 12-HOUR TIME INPUT: index.html birth-time is now HH(1-12):MM + an AM/PM (#birth-ampm); the submit handler converts 12h+AM/PM -> 24h for the engine (PM+!=12 -> +12, AM+12 -> 0) and keeps a birthTimeDisplay. PUSH to go live: cd /Users/jeffreyparker/CLAUDE/AFFIRMOLOGY/affirmology-site && rm -f .git/*.lock && git add index.html primer.html result.html && git commit -m "demo flow fixes" && git push origin master. OPEN: heading "One voice. One chart. One healing." may be replaced (Jeff weighing options).

  RESULTS PAGE REBUILT INTO THE PERMANENT HUB (2026-06-29, from Jeff's full-watch punch-list Affirmology_ResultsPage_PunchList_v1.md, NOT yet pushed). result.html #afterlisten now: removed the deep-report download + on-page blueprint glimpse (they only get the Soul Song; deeper report comes by email tomorrow); audio (soul song) download kept + centered; reframed post-song copy to give CONTEXT (this one tuned to IDENTITY+ALIGNMENT, a mantra of recognition, a taste of the engine); "this page is yours to keep, we emailed you the link, come back to replay/download/rewatch + new audios coming" copy; reworked INVESTOR ask into a real CTA ("own a piece... from $5,000... we'd love to have you... let's talk"); removed the DUPLICATE Mission & Opportunities link (one card only); MERGED the standalone beta page into the bottom = "Join the beta / Founding Constellation" with a simple checkbox "Put me on the beta testing list" + "I'm in" (uses the name/email we already have) -> POST /api/beta-signup (to wire) + thank-you; Connections line; Calendly (jeffparkerlove/speakwithjeff) moved here, not squished; the Joe & Monika experience FILM added at the bottom to REWATCH; renamed skip button to "Skip ahead"; report renderers guarded (dead/safe). PERSISTENCE: on return, if localStorage has affirmology:done:, showReturning() skips the film/ritual and shows the finished page (replayable song + full hub) directly; revealAfterListen sets the done flag. CAVEAT (honest): localStorage persistence is PER-DEVICE, so if they open the emailed link on a DIFFERENT device than they took the demo on, it will not remember them (no name, runs the flow again). True cross-device "remember me" needs the backend /api/status to also return the person's name + a done state by runId (a follow-up). affirmologist.html is now SUPERSEDED by the in-page beta section (harmless to leave). Push: cd .../affirmology-site && rm -f .git/*.lock && git add index.html primer.html result.html && git commit -m "results hub rebuild" && git push origin master.

  DEMO EMAILS REWORKED (2026-06-29, Jeff): NO MP3 in the email. Both worker emails (affirmology-demo-site/worker/src/index.js) rewritten to short, branded (emerald/gold, inline-styled) HTML from hello@affirmology.ai. The delivery email ("Your Origin Soul Song is ready") drops the MP3 button + PDF + transcript and instead has ONE gold "Open your page" button to https://demo.affirmology.ai/result.html#<jobId> (their permanent results hub, no &film) + reiterates "over the next few days we will send a few more gifts coded to your blueprint, always on your page." The confirmation email ("being composed") trimmed of the wrong "10-20 min / MP3+PDF follow-up" promise. (The /api/delivered handler still requires mp3_url as the render-done signal; it is just not emailed.) Worker JS verified (node --check OK). DEPLOY: the worker deploys SEPARATELY from the Pages site, via wrangler. FLAG for Code/engine: ensure the engine's own auto-email (affirmology-studio api/emailer.py, which attaches/links the MP3) does NOT also fire for demo renders, or the user gets a second MP3 email. RESULTS PAGE JS verified parses clean after the hub rebuild (node --check). NAMING applied on the page: "Origin Soul Song" (the audio), "Join our Founding Constellation" (the beta circle), "The Inner Orbit" (the whole circle of investors+testers+collaborators); note this shifts from the earlier playbook tiers (Inner Orbit was the testers) - reconcile the playbook later. TWO DEPLOYS PENDING (Mac): site cd .../affirmology-site && rm -f .git/*.lock && git add index.html primer.html result.html && git commit -m "results hub + naming + Origin Soul Song" && git push origin master; worker cd .../affirmology-demo-site/worker && npx wrangler deploy.

  THIS IS THE SINGLE SOURCE OF TRUTH ACROSS ALL CHATS.

  Protocol for every chat, no exceptions: 1. READ THIS FILE FIRST before doing anything or making any claim about project state. 2. Trust this file over older session notes, over scattered docs, and over your own assumptions. Where this file and another doc disagree, this file wins, or you reconcile them and fix this file. 3. UPDATE THIS FILE LAST before the session ends: what changed, what is now true, what is the next action. Keep it short and current. Delete stale lines rather than appending contradictions. 4. This file is intentionally unversioned. It is living. Do not make a v2; edit this one. 5. DELIVERING FILES TO JEFF: always surface any file as a clickable card via the present_files tool so it opens in Claude desktop. NEVER tell Jeff to find or open a file by path or filename. He will not hunt for it, and raw .md/.txt open in a hard-to-read editor. He compares quickly via links. When the point is reading or comparing, present it as a link and prefer rendered formats (PDF or HTML) over raw markdown.

  Last updated: 2026-06-24 (Cowork, timing/launch session, NO code). SESSION (electional + launch planning): Jeff missed the June 20 LLC filing (no funds + Saturday). Routed the electional/timing question to Hermes live (bridge healthy, corpus ok). Council result: best LLC file date before the retrograde = THU JUN 26 morning before ~1pm EDT (Taurus Moon trines his Virgo Sun-Venus 8th-house axis, Mercury still direct); workable backup Sat Jun 28; STRONGEST overall = AUG 11-13 (Sun on his Vertex 21 Leo 7th, Jupiter building into Leo, Moon over PoF/Jupiter). Mercury retrograde verified Jun 29-Jul 23 2026 in Cancer (shadow clears ~Aug 5), sitting ON his natal Cancer Moon (6th). Guidance: cold investor outreach before Jun 29 or after Jul 23; WARM re-engagement + warm beta invites are favorable DURING the retrograde. Money squeeze = Saturn in Pisces opposing his Virgo Sun-Venus (8th) + Sun crossing natal Saturn-Mercury this week; eases Jul 24-Aug 13; big expansion bell early Nov (Sun conj natal Jupiter 10th). Move now: brief warm check-ins to each paused coaching deal before Jun 29. DELIVERABLES (project root): Affirmology_TimingAndLaunch_ActionPlan_v1.md (master plan: dates, Mercury messaging, travel/beta ops, audio), Affirmology_Kairos_ElectionalAgent_Spec_v1.md (PROPOSED new electional/timing oracle "Kairos", NOT built, awaiting Jeff's go), Affirmology_DaytimeAnchor_AudioScript_v1.md (looping daytime reassurance/wealth script), Affirmology_ElectionalTimingReading_2026-06-24.md (raw council reading). RENDER INFRA CORRECTION (Jeff-verified 2026-06-24): Jeff verified that the CLOUD renders audios laptop-independent (works with the laptop out of the loop). This supersedes the older "the whole render path runs on the laptop / only works while the laptop is awake" notes in sections 11 + 13 + the section-5 operational fact, for the STUDIO/cloud path. Net: beta testers can enroll + receive audios 24/7 without Jeff's laptop, so the travel week is NOT render-blocked. (The LOCKED DEMO laptop path is a separate matter; this correction is about the Studio/beta render infra.) NEXT: confirm whether the Hermes/agent funnel reading path is likewise cloud-served or still laptop (today's reading went through the laptop funnel macbook-pro-2:8443). OPEN: (a) Jeff to pick audio render path (RESOLVED: chart-driven engine render, ~25-30 min; render command + Code paste delivered); (b) Jeff to approve building Kairos electional agent. NEW HARD RULE ADDED TO CLAUDE.md (Jeff 2026-06-24): ALL audio, anywhere/anytime, MUST be rendered through OUR system (engine/council via funnel), NEVER a third-party TTS. AUDIO BUG FOUND: the Hermes render_sacred_audio tool does NOT honor a custom source_script, it regenerates its default natal/chart structure (verified: a custom "I am Jeff" daytime-anchor script came back as a 2nd-person natal reading, entry 5e779947...). So custom long-form affirmation tracks must currently render via the engine CLI --script-file on the Mac (reliable), OR render_sacred_audio needs a fix to actually pass source_script through. Render-ready files in project root: Affirmology_DaytimeAnchor_AudioScript_v2.md (the script: no warm-up, name woven in, all first person, smorgasbord of techniques) + Affirmology_DaytimeAnchor_RenderScript_x3.txt (same script x3 = ~40 min for a long background loop). LOCKED LAUNCH TIMELINE (2026-06-24, Jeff said "lock it in"): seven-beat electional sequence, full detail + caveats in Affirmology_LaunchTimeline_v1.md. Beats: (1) Jun 26 file the WY LLC (direct on wyobiz.wyo.gov, agent ADDRESS+consent only NOT their filing service; resolve the wyomingagents.com $163 or use a fresh RA; EIN+bank, money can come in July; NO SAFEs); (2) Jul 4 beta opens FREE (no payments/upgrades that day; beta users never pay, free + escalating incentives); (3) Aug 11-13 HARD commercial launch (app live, Meta ads, payment links, customer money) + open the angel raise ($5-50k chunks, goal ~150k); app can ship via an individual Apple acct then convert to LLC; (4) Sep 11 first in-person Miami exposure + sales test (Jeff's bday/solar return); (5) Sep 24-26 Ultimate Wellness Event (Faena MC gig, agreement in Affirmology_UltimateWellness_MC_Agreement_v1.md); (6) Nov 11 consciousness GATHERING not commerce (Scorpio Mercury Rx is an asset for depth; lock venue + promote BEFORE Oct 24; venue target = Palapa, Jeff's feelers out); (7) early Dec Art Basel OPTIONAL/LATER after an explosive UWE, join an existing event, keep it intimate. Verified Mercury Rx 2026: Jun 29-Jul 23 (Cancer) + Oct 24-Nov 13 (Scorpio). NOTE: caught + corrected a council factual slip (it claimed Mercury clear on 11/11; it is not). LLC FILED 2026-06-25, EFFECTIVE 2026-06-26 (DONE): AFFIRMOLOGY LLC filed by Jeff himself on the Wyoming SOS portal. Filing/Original ID = 2026-002015324, stamped FILED 06/25/2026 11:05AM with a DELAYED EFFECTIVE DATE of 06/26/2026 (the elected birthday). Manager-managed (Jeff's title on the form = MANAGER). Organizer Jeffrey Parker; contact email jeff@affirmology.ai. Registered agent = "Wyoming Registered Agent" (the commercial agent behind Rocky Mountain Registered Agent, $29/yr), 1621 Central Ave Ste M, Cheyenne WY 82001 (also used as the org/mailing/principal address for privacy). NOTE: the Articles + Certificate only print the FILED date 06/25 (WY never prints the delayed effective date on the cert; that is expected and does NOT mean the effective date failed). The delayed effective date was confirmed on Jeff's pre-submission review screen. To get the 06/26 effective date in writing: the SOS entity record once indexed, or call SOS Business Division (307) 777-7311 with the Filing ID. Cost: $29 agent + ~$102.40 state. NEXT (on/after 06/26): free EIN at irs.gov, then open the business bank account; money can come in. Operating agreement still to be drafted (manager-managed; Sol 25% fully vested, Colin TBD). LAUNCH PUSH 2026-06-28 (Sunday eve): COWORK->AGENTS BRIDGE CORRECTION (important): the live agents ARE cloud-native and healthy at https://affirmology-studio-api.onrender.com (GET /api/health ok, elevenlabs_key_set, R2 corpus+music bed); auth Bearer CLOUD_STUDIO_AUTH_TOKEN (in affirmology-studio/.env). The Tailscale funnel (macbook-pro-2...:8443) is RETIRED and only works when the laptop is awake -- do NOT use it from Cowork; use the onrender cloud endpoint. (A chat mistakenly used the funnel this session; cloud was up the whole time.) COUNCIL SEND-WINDOW RULING (cloud, June 28): Jeff's outreach list is relationship-WARM but Affirmology is a FIRST REVEAL to them = hybrid; Mercury Rx reads a first reveal as NEW, so it does NOT get the full warm pass. Best send window = TODAY June 28 (Mercury still direct); Monday June 29 morning before ~10am ET (before the early-afternoon station) for the rest; into the Rx only with the "private first look" framing ("I've been keeping this close, this feels like the right moment to bring you into it, I'd love your first-read reaction before we go wider"). Pluto square his MC at 0.03 orb = live career-destiny pressure, move now. PLAN AS OF TONIGHT: Jeff finishing the investor/movement video in CapCut (Palmier MCP timed out, abandoned for now; can revisit to fill gaps later), testing the full demo flow; rest of the deliverable done; sending Instagram DMs tonight (socially ok late) + texts/WhatsApp Mon 8-10am. THREE PRODUCT NOTES (Jeff): (1) the per-person Report/Reading email delivers NEXT-DAY (delayed) after demo access, set that expectation in the message; (2) onboarding alerts that a NEW custom gift audio drops each day; (3) apply for the ElevenLabs freebie once the demo is up. Design help: Affirmology_DesignBrief_ForAnyChat_v1.md (paste into any chat so output stops being generic). NDA tool pick = SignWell (embedded in-browser signing, gates access, tracks signers, ~$10/mo; BoldSign pay-per-use alt; NOT Dropbox Sign for embed). PRIOR SESSION BELOW. --- Earlier 2026-06-21 (Cowork, C-build session). SESSION (C29 backfill + C28 confirm + C26 STEP 3 BUILT): Audited the four-step app-changes build and found STEP 1 C29 (rule+form+scope craft schema) and STEP 2 C28 (practitioners/wisdom nightly research lane) were ALREADY built + committed (studio commits 347bc09 and 3975330 + the practitioners chain a21fbcf/e922b6e/8ce25f8/612797c); verified green (test_craft_rule_form.py + test_practitioners.py). Ran the C29 craft backfill --apply on the live corpus (corpus/corpus.db): 112 craft entries, 106 already tagged, the last 6 processed for $0.0038 (5 production/QA-craft entries correctly left untagged, 1 tagged rule=7 form=journaling_prompt,mindset_practice). BUILT + COMMITTED STEP 3 C26 (audio delivery pacing + activation), studio commit 601e7f0, Studio-only + additive + demo byte-for-byte unaffected (standard pace is a true no-op): Structure gains pace + bed_profile (before_bed + gk_single_sphere = deep; confidence_abundance = activation/driving; natal_sacred stays standard); new agents/pacing.py (pace->speed deep 0.88 / standard 0.95 / activation 1.03 under the 1.05 ceiling; mix-time silence after questions + at section breaks via new audio_mix.concat_voice_segments; synthesize_voice_paced segmented synth, standard delegates to the unchanged single-call path; target_word_count for length consistency; deterministic pacing_gate); cli.py gains --pace / --bed-profile and a paced QC-gated synthesis branch (QC gate + 1.05 ceiling preserved); council.compose takes pace/bed_profile (Hypnos sequences pacing, Orpheus writes the question + laddered arc, Apollo folds in the pacing gate); studio pipeline.py + jobs.py thread pace from the structure (a standard render still passes --speed 1.0, no --pace). Engine changes vendored into affirmology-studio/affirmology-agent (kept in sync). Tests: test_c26_pacing.py (4 acceptance, decisions + CLI args) + test_c26_pacing_render.py (real ffmpeg silence insertion passes the QC gate; QC still catches a true dropout), 6/6 green; engine pytest 46/46; full studio suite 114 pass / 1 fail = test_chart_cache::test_determinism_cached_vs_fresh, a PRE-EXISTING aspect-ordering nondeterminism proven independent of this work by a baseline run (stash-all -> still fails), unrelated to C26 and a chart-engine fix for later WITH Jeff. STOPPED before any paid ElevenLabs render per Jeff: NEXT = Jeff listen-checks one deep render (before_bed) + one activation render (confidence_abundance), then STEP 4 (C3 tones). Queue updated: C29 schema + C28 checked off, C26 marked [~] built/pending-listen-check, C3 left for after the listen-check. EARLIER SAME DAY: reconciled the Test & Change Queue (C22/C24/C25 were shipped 2026-06-20 but still showed unchecked; now checked off, PROJECT_STATE remains the source of truth). Wrote four new build docs: Affirmology_AudioPacingActivation_C26_BuildBrief_v1.md (slower deep delivery, pauses, activation questions, per-Structure pace/target_minutes/bed_profile, render-length consistency), Affirmology_FrequencySoundLayer_C3_Spec_v1.md (synthesized isochronic/binaural/tonal entrainment layer in audio_mix.py, state-to-frequency map, honesty rule, off by default), Affirmology_BeliefInstallation_C29_Kickoff_v1.md (the rule+form tagging schema for the craft/practitioners corpus, plus first wealth-journey journaling prompts + mindset practices), and Affirmology_SoundDesign_DynamicBeds_Report_v1.md (plain-language sound report: layers/stems/sections, vocabulary, do-now-vs-hire-later, Suno recipes for Sol, free/public-domain learning). Also two prompts: Affirmology_AudioStructures_CoworkSession_StarterPrompt_v1.md (launch a Cowork session to organize techniques + design structures/suites: starter kit, transformation tracks, wealth suite with affirmations/journaling/EFT/before-bed/motivational) and Affirmology_AppChanges_MasterCodePrompt_v1.md (one sequenced Code paste: C29 schema -> C28 research lane -> C26 pacing -> C3 tones, demo locked, QC kept, vendored). Added a free/public-domain sources section to the C28 plan (Neville/Wattles/Shinn/Troward/Behrend/Allen/Coue/Kybalion ingest-fully; in-copyright + RTT benchmark-only; hypnotherapy researched first). NOTE: the dedicated C28 hypnotherapy/practitioners research lane is NOT built yet; usable hypnosis craft today lives in the craft areas meditation_hypnosis (15) + breakthrough (24). NEXT for Jeff: (a) commit the test-only C24 acceptance test, (b) paste the master Code prompt or run C26 first, (c) launch the audio-structures Cowork session, (d) give C28 the add-all/trim. Prior 2026-06-20: HERMES IS NOW EFFECTIVE AND PROVEN LIVE. The agentic loop fix + two hardening fixes are deployed on studio.affirmology.ai (commits ca3a055/eb6c075): FIX A front-end polls /api/hermes/history so there is no more "could not reach my model" flash on a slow read; FIX B keeps tools=HERMES_TOOLS in the forced-synthesis call so heavy multi-oracle reads no longer 400. T1 PASSED: a live two-person (Jeff+Sol) wealth-codes reading returned a full mechanism-level answer (Pearls 45.5/49.5, channels, Lilith/Chiron blocks, live transits), and a craft-grounded sleep-technique reading fused chart + craft + timing, brand-clean (heart coherence breath, never the teacher's name). Craft Knowledge Engine Phase 2 is wired AND live on cloud (craft + techniques synced to R2, commits 5be790e/bc507c1; cloud /api/corpus-stats: craft=94, techniques=18, records=25570). NEW DIRECTION (Jeff): HERMES AS COMMAND CENTER, "a cheat code superhighway into all our systems," Jeff full keyset, Sol/Colin governed. Map: Affirmology_Hermes_CommandCenter_CapabilityMap_v1.md. The shared ACTION LAYER (endpoint + Hermes tool + chat button + server-side tier check) is GREENLIT with briefs written: C16 cross-chat messaging (Affirmology_CrossChatMessaging_C16_BuildBrief_v1.md) + C17 send-to-Studio (Affirmology_SendToStudio_C17_BuildBrief_v1.md), to BUILD TOGETHER in one Code pass (combined paste delivered in chat). C15 field-briefing / read-aloud audio rides on that layer (Affirmology_FieldBriefingAudio_C15_BuildBrief_v1.md + cost memo Affirmology_FieldBriefingAudio_CostMemo_v1.md): two modes, casual read-aloud is primary ("read me the chat while I cook"), field-briefing-with-save is the upgrade. ENGINE PATH for C15: start on ElevenLabs (already wired, no new account), HARD-capped per user + tier-gated server-side (Jeff's cost constraint), then move the default read-aloud voice to a LOCAL open TTS (Kokoro/Piper on the Mac mini) for ~$0; do NOT rely on OpenAI (Jeff cancelled it); Fish Audio is the cheap hosted fallback. Public share LINK for saved audio still needs C10 (R2 public media). Single living backlog of tests + changes: Affirmology_TestAndChange_Queue_v1.md. PENDING NEXT: paste the combined C16/C17 build to the studio Code window; the local funnel server still needs a manual restart to pick up the Phase 2 engine changes for local --council renders. Also logged: Hermes chat readings do NOT pass Apollo's craft gate, so chat-side accuracy slips (e.g. one reading briefly conflated Pearl 44.1 with 45.5) need a lightweight Hermes self-check, to be folded into Phase 3 of the craft grower.

  Earlier 2026-06-18 (Cowork, Atlas+Studio planning session: Jeff's brain dump captured. ATLAS DECISIONS LOCKED (section 17b): in-app messaging Jeff/Sol/Colin only, Hermes takes actions, permission tiers, remember-last-mode, transcript-only voicenotes, nightly auto-applies SAFE changes only, Whisper now / Parakeet later, brand the Cloudflare Access page. STUDIO UPGRADE QUEUE added (Affirmology_StudioChangeList_v1.md top block): numerology full-name accuracy guard, structure overview/scope panel, length up to 45min, music-bed selector (None+Heavenly now, pace categories defined), conversation history, demo-candidate labeling, openings/endings research, slower ElevenLabs pacing, relationship in-app, internal-report transparency. NEW COUNCIL AGENT: PERSEPHONE, the breakthrough/transformation guardian (section 12). Prior entry below: (1) CHIRON, the Technique Healer, added to the council as a new prompted role between Sophia and Orpheus, with a before/after on Jeff's real chart. See section 12. (2) LAPTOP HOUSEKEEPING now that the Studio is on the cloud: retired the local laptop Studio server, moved the corpus OFF the external SSD to an internal path under the AFFIRMOLOGY folder, and corrected the misleading corpus status counting. See section 20. Earlier same day: STUDIO DEPLOYED TO THE CLOUD on Render + R2, full render pipeline proven end-to-end incl. PDF; corpus + music bed pulled from R2 on boot; auth locked; dependency audit. See section 19. Also: corpus-structurer model bake-off = HOLD on Haiku, and the "remaining backlog" is non-structurable junk, both in Affirmology_Nemotron_CorpusUpgrade_Plan_v1.md.)

  ACTIVE WORK POINTER (Studio + Atlas build, 2026-06-18): the work order for today's Claude Code session is Affirmology_ClaudeCode_BuildBrief_2026-06-18_v1.md (Studio upgrades S1-S10, then Atlas v1 B1-B9, with exact files, acceptance criteria, and guardrails). Build Studio first (Sol tests tonight), then Atlas. Persephone gate prompt: CLAUDE OUTPUTS/Affirmology/Affirmology_Persephone_GatePrompt_v1.md.

  ACTIVE WORK POINTER: for the current website-video and demo-web redesign haul, READ Affirmology_SessionHandoff_VideoBuild_v1.md next. It is the portable continuity capsule with the locked cosmos-film direction, the generated Lily voiceovers and their word-timing maps, the tech approach, and every open task. Use the design-motion-principles and emil-design-eng skills on the film.

  ---

  0. The continuity rule (why this file exists)

  Past chats kept starting ignorant and repeating or undoing work. Concrete failures that prove it: session notes claimed the transits tradition had 0 docs when the database actually had 146; one chat did not know the demo audio baseline another had set; a chat claimed the corpus was not architected into generation when it was. The cure is this file plus the discipline above. If you are a chat reading this, the most useful thing you can do is keep it accurate.

  ---

  1. State at a glance

  • Corpus: built and largely structured. ~10,600+ structured records across 6 traditions. Details in section 3.
  • DEMO AUDIO LOCKED + FIXED (2026-06-15): the warble/dropout problem is solved. The demo now renders at the proven-clean baseline speed 1.0 (no aggressive auto-tune upward; the global auto-tune ceiling is capped at 1.05), and scripts/audio_qc.py runs on the voice-only track of EVERY render and refuses to serve a dropout (re-renders slower first). WeasyPrint was missing (that is why demo PDFs never generated); it is now installed and the macOS lib path is wired, so the PDF generates. Re-rendered Jeff at 1.0: QC PASS, clean, 6:17, PDF OK. THE DEMO IS LOCKED; all experiments live in the Studio. See sections 4 and 5.
  • CLOUD STUDIO BUILT + VERIFIED (2026-06-15): new affirmology-studio/ app, cloud-shaped per Affirmology_CloudStudio_BuildSpec_v1.md, running local-first on this laptop off the agent's keys. Two-phase, structure-driven, script-first cost discipline. Verified end to end. See section 11.
  • STUDIO LIVE (2026-06-16): studio.affirmology.ai is up and gated (Cloudflare Access, email allowlist for Jeff + Sol), served by the affirmology-studio-api Worker to the Tailscale funnel to the laptop; media persists on the SSD. Jeff rendered his own audio and it "sounds like my story." The single build queue + live-use bug list is Affirmology_StudioChangeList_v1.md. CRITICAL open bug: chart-integrity (a placeholder chart leaked into a report). Side door still open: disable the workers.dev URL. See sections 11-13.
  • Demo audio quality: baseline is locked (Jeff Parker render, Charlotte voice). See section 4.
  • Demo backend: LIVE end to end as of 2026-06-13. Worker deployed (route demo.affirmology.ai/api/*) with secrets, FastAPI server running on the laptop, exposed via Tailscale Funnel at https://macbook-pro-2.tail333640.ts.net . Chain proven externally (browser to Worker to laptop and back). Real form-submission render test is the next step. See section 5.
  • Corpus-fed generation: WIRED 2026-06-14 (pending a live test render). script_generator.py now queries the corpus for the chart's spine and grounds the script in those real snippets; the v5 reference is now a content-free structural skeleton so no founder data can leak. See section 2.
  • Mac mini: still paused (USB hub replacement). Demo backend lives on the laptop for now.

  ---

  2. What is wired vs not (read before claiming anything about the pipeline)

  WIRED and working: - Chart engines (astrology, Human Design, Gene Keys, numerology), verified against Jeff, Josh Parini, Sol fixtures (pytest). - Script generation, voice synthesis (ElevenLabs Charlotte), FFmpeg mix, Sacred Audio PDF. One CLI command produces a full render. - CLI Studio flags (2026-06-15): --concept "<directive>" (the structure directive; passes through to generate_script(concept=), content stays chart-only) and --build-report (writes the team-facing backend/build_report.md: design rationale, what was used/why, what was NOT used/why, suggestions, and a techniques footer from the Techniques Library). New module agents/build_report.py. Cost discipline: the script uses the strong model (Sonnet), the script map and build report use Haiku (AFFIRMOLOGY_BACKOFFICE_MODEL, default claude-haiku-4-5). - Audio QC gate (scripts/audio_qc.py) is NOW WIRED into the CLI --audio path (2026-06-15): it runs on the voice-only track after synthesis, and on a detected dropout it re-renders at a lower speed (--qc-retries, default 2) and refuses to serve a bad file (no mp3/PDF produced). It exposes run_qc() for in-process use. Speed: --speed is honored; auto-tune ceiling lowered to 1.05 so it never warbles. - Corpus scrape + structure pipeline (Haiku backend), with whole-document chunking and a re-mine mode. - Verifier that regex-checks scripts against the chart JSON to catch hallucinated placements.

  NOT wired yet (do not claim otherwise): - Corpus-fed generation: WIRED 2026-06-14 (pending a live test render). script_generator.py now calls _corpus_sources(chart), which uses db.search_records() to pull real A/B-tier interpretation snippets for the chart's spine (the four Gene Keys spheres, Sun/Moon/Rising, HD type and authority, Life Path) and injects them into the prompt as SOURCE MATERIAL so the language is grounded in the tradition texts. The v5 reference script was replaced with a content-free STRUCTURAL SKELETON, so no founder placements can leak into anyone's script. Corpus DB path is env AFFIRMOLOGY_CORPUS_DB (default /Volumes/Affirmology/corpus/corpus.db); if the DB is missing or locked the retrieval returns empty and generation proceeds chart-only (robust, will not break a render). STILL TO VERIFY: run a real render and read the script to confirm quality and that the grounding shows through. The fuller PersonalizationAgent design (context bundle, specialist) is in Affirmology_SubconsciousOS_AgenticArchitecture_v1.md Part Four. - Demo web backend (server + Worker) not yet stood up.

  ---

  3. Corpus state (MOVED 2026-06-18 to the internal path /Users/jeffreyparker/CLAUDE/AFFIRMOLOGY/corpus; was on the external SSD. See section 20 for the move + the corrected structurable status.)

  Scorecard (docs / words / structured docs): - western_astrology: 550 / 3.61M / 34 - human_design: 1,487 / 1.57M / 706 - transits: 382 / 1.08M / 369 - vedic_astrology: 527 / 0.88M / 42 - gene_keys: 413 / 0.40M / 186 - numerology: 507 / 0.19M / 7 - Total: ~3,866 docs, ~7.7M words, ~1,344 structured docs. - Live record counts (2026-06-14, SELECT tradition, COUNT(*) FROM record): human_design 7,863; transits 2,525; gene_keys 1,524; western_astrology 544; vedic_astrology 362; numerology 87. Total ~12,905 records, up from ~10,600 on 2026-06-13. Doc/word/structured-doc columns above are the 2026-06-13 snapshot.

  Important nuance: the low "structured %" on western, vedic, numerology is NOT a budget wall. Those traditions are big scanned books. Their real content is partly mined; most of their unstructured docs are sub-100-word junk pages that are correctly skipped. The genuine remaining work is (a) finishing the human design YouTube backlog, and (b) re-mining the big books deeper with --redo-min-words plus a higher --max-chunks-per-doc (needs more credit).

  Code capabilities added this week (in affirmology-agent/src/affirmology/corpus/): - Haiku/Anthropic structurer backend (default), removes the Gemini/Google billing blocker. --backend anthropic. - Real structure-only mode that structures existing extracted docs. - Whole-document chunking (--chunk-chars, --max-chunks-per-doc) so big books are read past the front matter. - Re-mine mode (--redo-min-words) that clears a long doc's shallow records and re-chunks it deeper, no duplicates. - Stale-path remap (_resolve_extracted_path) so paths from old sandbox sessions resolve to the SSD. - transits added to the default tradition set; transit element types added to the structurer prompt. - New gene keys sources added; human design YouTube channels switched to uncapped. - scripts/overnight_corpus.sh: chained, unattended, cost-capped corpus run with a morning report.

  NEW TRADITION (2026-06-19): techniques folded into the corpus. The Techniques Library (Affirmology_TechniquesLibrary_v1.md) is now structured into 18 tier-A records (inductions, identity/language, somatic, state/timing, sound layers, closing, plus the 3 proposed/benchmark techniques) under tradition techniques, so the engine and the oracles (Chiron especially) retrieve methods the same way they retrieve chart interpretations. Ingest is the re-runnable module affirmology-agent/src/affirmology/corpus/techniques.py (python -m affirmology.corpus.techniques), idempotent (wipes prior techniques records first), so re-run it when Jeff's heart-coherence technique is added. techniques added to _TRADITIONS in affirmology-studio/api/hermes_tools.py (the list search_corpus iterates and exposes in its tool enum) and the tool description now points the council at the methods layer. Verified live: /api/corpus-stats reports techniques: 18 after a server restart. NOT scraped, so it is intentionally left out of the scraper's run.py TRADITIONS list.

  CRAFT KNOWLEDGE ENGINE - PHASE 1 DONE (2026-06-19, per Affirmology_CraftKnowledgeEngine_Plan_v1.md). The second knowledge pillar (craft = HOW we say it and make it land, vs the chart = WHAT we say). New distilled store: table craft on corpus.db, owned by affirmology-agent/src/affirmology/corpus/craft.py (CraftStore, search_craft with tokenized relevance ranking, search_craft_text helper). Each entry is actionable: area / method / why / when_to_use / how_to_apply / one_liner / tags, across five areas (neuroscience, frequency_sound, audio_production, meditation_hypnosis, reading_craft). SEEDED by a one-time Haiku distillation (affirmology-agent/src/affirmology/corpus/craft_seed.py, reusable by the Phase 3 nightly grower) from 6 docs: TechniquesLibrary, Neuroscience_AudioEfficacy, PredictiveBrain_IdentityChange, AudioMastering_Production, DepthPsychology_JungianIntegration, and the ReadingCraft standard. Result: 94 entries (audio_production 24, reading_craft 38, meditation_hypnosis 16, neuroscience 9, frequency_sound 7), cost $0.19 of a $1.50 cap. Brand-checked (no em dashes, no named teacher, no medical claims). Idempotent: re-run python -m affirmology.corpus.craft_seed to re-seed (clears each doc's prior seed rows first). STILL TO DO - Phase 2: wire it in (replace the static TECHNIQUES_LIBRARY menu in build_report.py with craft retrieval; give Chiron/Orpheus a search_craft tool; add a craft check to Apollo's efficacy gate; add search_craft to Hermes in affirmology-studio/api/hermes_tools.py). Phase 3: extend the nightly growth cron to grow the craft tradition (crawl wide, distill, fold in feedback). Phase 4 (later): frequency/sound overlays in audio_mix.py (Studio-only, demo untouched).

  CRAFT KNOWLEDGE ENGINE - PHASE 2 DONE (2026-06-19): the craft base is now wired into the agents (Studio --council path only; the LOCKED DEMO never passes --council and is untouched, and generate_script is unchanged). Ownership per the corrected plan: HYPNOS is the PRIMARY owner of technique selection/sequencing. Changes: (1) affirmology-agent/src/affirmology/agents/council.py - new select_craft() (technique set + a healing slice for CHIRON), writing_craft() (reading_craft slice for Orpheus), hypnos_plan() (Hypnos selects/sequences an ordered technique plan from the live craft base), and HYPNOS_SYSTEM prompt; compose() runs Hypnos once, injects the plan + writing-craft into Orpheus's draft context, and feeds the craft plan + base into apollo_review() (new craft_plan/craft_base args) so Apollo's technique/hypnos gates check the script against the craft base; the Hypnos plan is written into council_transcript.md. (2) affirmology-agent/src/affirmology/agents/build_report.py - the static TECHNIQUES_LIBRARY menu is REPLACED by live search_craft retrieval (_craft_menu_text); the static list is kept only as a fallback. (3) affirmology-studio/api/hermes_tools.py - new search_craft(query, area) tool (function + schema + dispatch) so chat readings ground in craft. (4) affirmology-studio/api/main.py - Hermes system prompt documents the search_craft tool. ALL craft retrieval degrades to empty on any failure, so a render never breaks. Verified: imports clean, craft menu/plan/tool all return real entries, demo path untouched, graceful fallback confirmed. REQUIRES a git push of affirmology-studio (for the Hermes search_craft tool to reach studio.affirmology.ai) AND a restart of the local funnel server (so the live agents + the --council render path pick up the engine changes). Phase 3 next: extend the nightly growth cron to grow the craft tradition.

  DEPLOYED TO CLOUD 2026-06-20 (commits ca3a055 + eb6c075, live on studio.affirmology.ai): the Phase 2 craft wiring + two Hermes hardening fixes shipped together. (a) Hermes FIX A (front-end polling in web/app.js: hermesSend polls /api/hermes/history for up to 3 min instead of flashing an error on a slow/throwing POST; cache-bust v18->v19). (b) Hermes FIX B (api/main.py: both forced-final-synthesis calls now keep tools=HERMES_TOOLS, fixing the multi-oracle BadRequestError; reading quality + oracle model untouched). Both per CLAUDE OUTPUTS/Affirmology/Affirmology_HermesLoop_Bug_v1.md. VENDORED-ENGINE SYNC: the cloud Dockerfile bundles affirmology-studio/affirmology-agent/ (AFFIRMOLOGY_AGENT_ROOT=/app/affirmology-agent), which was missing corpus/craft.py and had stale council.py/build_report.py; all three were synced into the vendored copy so cloud --council renders match local. CRAFT DATA SYNCED TO R2 + LIVE ON CLOUD 2026-06-20 (commit 5be790e). MERGED, not overwritten: pulled R2 corpus/corpus.db (the studio boot key), confirmed the nightly cron had NOT run (R2 had no meaning/distinctions/craft tables; record=25552 = the 6 original traditions), backed the original up to corpus/backups/corpus.db.pre-craft-20260620, copied the craft table (schema + all 94 rows) into the R2 copy via ATTACH (integrity_check ok; record/source/document unchanged at 25552/102/3880), and pushed it back to the same key (18,796,544 bytes). corpus_boot re-pulls on size mismatch, so the redeploy reloaded it. VERIFIED LIVE: cloud /api/corpus-stats returns craft=94 (audio_production 24, reading_craft 38, meditation_hypnosis 16, neuroscience 9, frequency_sound 7), records still 25552, size 18796544. /api/corpus-stats was extended to report craft/meaning/distinctions counts (public). NOTE: the authenticated cloud Hermes search_craft call could not be run from here (cloud STUDIO_AUTH_TOKEN is Render-only + Cloudflare Access); verified equivalently via search_craft against the exact merged DB (real entries, not fallback). TECHNIQUES TRADITION ALSO SYNCED TO R2 + LIVE 2026-06-20 (empty redeploy bc507c1): same merge method, copied the techniques source+document+18 record rows into the R2 corpus (backup corpus/backups/corpus.db.pre-techniques-20260620; no id collisions; integrity_check ok, foreign_key_check clean; craft stayed 94 and all 6 base traditions unchanged), pushed back (18,808,832 bytes), cloud re-pulled. Cloud /api/corpus-stats now: records 25570, by_tradition includes techniques=18, craft=94. Cloud corpus now matches local. Local funnel server still needs a manual restart to pick up the Phase 2 engine changes for local renders.

  C16 + C17 BUILT + DEPLOYED 2026-06-20 (commit 96d8dd6, live on studio.affirmology.ai; app.js v20). Per Affirmology_CrossChatMessaging_C16_BuildBrief_v1.md and Affirmology_SendToStudio_C17_BuildBrief_v1.md. Shared shape (bearer endpoint + permission-gated Hermes tool + chat button); api/ + web/ only (no engine change, demo untouched). C16 cross-chat messaging: deliveries table in jobs.py (add_delivery/list_inbox/mark_delivery_read, read scoped to recipient); POST /api/hermes/push, GET /api/hermes/inbox, POST /api/hermes/inbox/read; send_to_person Hermes tool; founder-circle tiers (jeff/sol/colin) enforced SERVER-SIDE in hermes_tools.check_send_permission (josh rejected); web "Send to..." control + inbox surface. C17 send-to-Studio: draft job state + source/music_bed/production_notes/draft_script columns (migrated) + create_draft() + JobManager.render_draft() (promotes a draft through the normal chart-driven script path on the same entry); POST /api/studio/draft (status draft, source hermes), POST /api/studio/draft/{id}/render; send_to_studio tool; web "Send to Studio" control + Studio-home drafts tagged "from Hermes" with a Render button. Both tools are USER-INITIATED ONLY (descriptions forbid autonomous sends) and ADDITIVE ONLY. Chart-driven preserved: the Studio stays canonical; the draft script is editable/regenerable. All 8 acceptance checks pass at API/tool level (incl. permission rejections); verified live by route gating (401 = registered). LATER (per briefs): audio payloads (C15), notifications, read receipts, advisor<->client; render-as-is vs regenerate nicety; Hermes-triggered render. FOLLOW-UP FIX 2026-06-20 (commit 849d9e1, live): the action layer shipped in 96d8dd6 but the Hermes system prompt never listed send_to_person/send_to_studio, so Hermes self-reported it "cannot" push to a person or create a Studio draft and would not use them. Added both tools to _hermes_system with the user-initiated-only framing ("you DO have this ability; only act when the user explicitly asks"). Added BUILD_TAG to /api/health so prompt-only deploys are verifiable. If Hermes is tested on the LOCAL funnel server, restart it to pick up all of this (and the still-pending Phase 2 engine changes).

  C18 HERMES-RENDERED SACRED AUDIO BUILT + DEPLOYED 2026-06-20 (commit 4a0e0ea, live; build tag 2026-06-20-c18-hermes-rendered-sacred-audio; app.js v21). Per Affirmology_HermesRenderedSacredAudio_C18_BuildBrief_v1.md. Completes the action layer: Hermes renders a full Sacred Audio (voice + PDF) from chat through the EXISTING proven render/QC/PDF pipeline, delivers the finished player + PDF into the chat, then offers to share it to a searchable library. api/ + web/ only (no engine change, demo + render settings untouched); user-initiated + additive + tier-checked server-side. STORE (jobs.py): deliveries.pdf_ref + jobs.visibility(default private)/created_by (migrated); set_visibility() + list_shared(q over title/person/structure, done+shared only); JobManager.render_sacred_audio()+_run_full_render (reuses _run_script then _run_audio, then drops a kind "audio" delivery from "hermes" to the requester carrying mp3_url+pdf_url; non-blocking single worker). TOOLS (hermes_tools.py): render_sacred_audio (creates the entry via C17 create_studio_draft internals, starts the render, returns immediately with entry_id + a "rendering" message), check_my_render (status -> media urls when done), share_studio_entry (owner-only: jeff any, sol/colin own via created_by), search_studio_library; check_share_permission; run_tool now also receives the manager. ENDPOINTS (main.py): POST /api/studio/entry/{id}/share, GET /api/studio/library?shared=1&q=; inbox surfaces pdf_ref; all four C18 tools added to the Hermes SYSTEM PROMPT with the honest "everyone with Studio access = the Cloudflare Access allowlist" wording. WEB: Hermes inbox renders kind "audio" deliveries as an inline audio player + PDF link; Studio home has a Shared Library list + text search (loadSharedLibrary, GET /api/studio/library). All C18 acceptance checks pass at the wiring level (render kickoff, completion delivery with media, check tool, share + search, cross-owner permission block); the full ElevenLabs render was NOT run in the test to avoid spend (it uses the unchanged proven pipeline). LATER (per brief): customer-facing name (Cast/Invoke/Conjure/Summon) TBD; render-as-is vs regenerate.

  C18 PART 1 (URGENT FIX) + PART 2 BUILT + DEPLOYED 2026-06-20 (commit 788c358, live; build tag 2026-06-20-c18-part2-cast-deliver-to; app.js v22). PART 1: Hermes 400'd on follow-up turns ("check on it") because hermes() built the model messages from the last 24 stored turns, a window that could start with an assistant turn or contain consecutive same-role turns (a reply + a render-completion delivery), both of which Anthropic rejects. Fix: _normalize_history() in api/main.py drops empty turns, drops leading assistant turns, and merges consecutive same-role turns so roles strictly alternate (falls back to the current user message if the window collapses); the except block now logs full str(exc)+traceback. Unit tests in affirmology-studio/tests/test_hermes_history.py (5/5). PART 2: (1) permissions relaxed to Studio-openness WITHIN the circle: any of jeff/sol/colin may render from any member's chart, deliver to any member, and share/unshare any circle record (check_draft_permission now requires sender+subject in circle; check_share_permission requires sender in circle). Hard stops only: reject anyone outside the circle (actor/subject/recipient); NO delete path exists (verified); demo locked. (2) render_sacred_audio gained deliver_to with REGENERATE-PER-RECIPIENT: a Cast to X is regenerated from X's OWN chart (person=X) and delivered to X, never a forward; EXCEPTION: relationship/teamwork structures (new Structure.is_relationship flag + is_relationship_structure() keyword detector in structures.py) are ONE shared audio delivered to both, no regen. JobManager.render_sacred_audio/_run_full_render now deliver to a list of recipients; created_by records the actual caster. (3) UI Hermes-render button labeled "Cast" (tool stays render_sacred_audio). (4) start_public.sh intentional edit committed (media defaults to ./data/media, no SSD needed; token print masked to last 4 chars, no secret committed). api/ + web/ only (no engine change, demo untouched), server-side enforcement. All PART 1 + PART 2 acceptance checks pass at the wiring level (full ElevenLabs render not run in tests, to avoid spend).

  C19 STORED NATAL CHART BASIS + TIMING SNAPSHOT BUILT + DEPLOYED 2026-06-20 (commit 8452cda, live; build tag 2026-06-20-c19-stored-natal-basis). Per Affirmology_StoredNatalChartBasis_C19_BuildBrief_v1.md (two-tier). TIER 1 (immutable natal basis, compute once + reuse): the FULL natal Chart is cached and reused for all audio gen, Hermes, and readings. Vedic is now part of the basis: added a vedic dict field to the Chart model (api models.py), populated once in compute_chart (derived from tropical longitudes), so the complete blueprint (western incl. aspects/extended points/asteroids, HD, Gene Keys both sequences + Pearl, numerology, AND Vedic) serializes + round-trips faithfully. New affirmology/chart/cache.py: CHART_ENGINE_VERSION="2026-06-20.1" (bumping busts all caches so engine upgrades recompute once), chart_cache_key=sha1(normalized birth + version), compute_chart_cached/get_or_compute. Lookup order: committed fixtures -> SQLite chart_cache (env AFFIRMOLOGY_CHART_CACHE_DB, set to studio.db) -> compute+store. Committed natal fixtures for Jeff/Sol/Colin/Josh in src/affirmology/verification/known_charts_computed/, regenerable by scripts/precompute_known_charts.py on a version bump (instant on laptop + cloud, zero first-compute). TIER 2 (timing snapshot, date-dependent, layered on the basis): progressed_summary (progressed Sun, progressed Moon + lunation phase, upcoming sign changes) + timing_snapshot in chart/timing.py; timing_snapshot_cached per (person, date) with a TTL in cache.py. render_timing_facts STAYS LIVE (date-stamped) so timing always reflects now. WIRING: compute_chart_cached routed through get_person_chart/compute_chart_from_birth (hermes_tools._compute), people.chart_summary, overview, and the render path (cli.py) + orchestrator; studio config exports AFFIRMOLOGY_CHART_CACHE_DB=studio.db so the engine CLI subprocess shares the cache. Guardrails: identical output (caching is pure performance; natal facts byte-identical cached vs fresh, proven), demo render settings unchanged, no em dashes, engine changes VENDORED into affirmology-studio/affirmology-agent/ (verified the cloud copy loads fixtures with 0 compute). Tests: C19 acceptance 8/8 (tests/test_chart_cache.py), engine pytest 46/46. NOTE: acceptance #5 proves the natal BASIS/facts are byte-identical cached vs fresh (the audio's inputs are unchanged); a literally byte-identical audio FILE also depends on the non-deterministic LLM script step + TTS. ON VERSION BUMP: re-run scripts/precompute_known_charts.py and re-vendor the fixtures.

  HERMES LONG-THREAD 400 FULLY FIXED 2026-06-20 (commit fbf49e9, live; build tag 2026-06-20-hermes-history-endonuser-fix). The C18 PART 1 normalize fix did NOT fully resolve it; reproduced the real error with a live API call: "This model does not support assistant message prefill. The conversation must end with a user message." Root cause: list_hermes_history used ORDER BY created_at ASC LIMIT 24 = the OLDEST 24 turns, so on a long thread it dropped the just-added current message and the window ended on an old assistant turn (400). Locally it never showed because jeff had only 18 turns. Fixes: (1) list_hermes_history now returns the most recent N turns in chronological order (subquery DESC LIMIT N re-sorted ASC) so the current message is always last; (2) _normalize_history also drops TRAILING assistant turns so the window always ends on a user turn. Verified by replaying real stored history (+a simulated >24-turn thread) through a live Anthropic call: succeeds (stop=end_turn) instead of 400. Unit tests 6/6. api/ only (jobs.py + main.py), no engine change, no vendor.

  DAILY-DIGEST EMAIL ENDPOINT ADDED + DEPLOYED 2026-06-20 (commit 94f16be, live; build tag 2026-06-20-send-digest-resend). POST /api/send-digest (bearer-auth, studio token; for a scheduled task) sends a report email via the Resend REST API (POST https://api.resend.com/emails, Authorization: Bearer RESEND_API_KEY) from FROM_EMAIL (default reports@affirmology.ai on the verified affirmology.ai domain). Body {subject, html, text?, to:[emails]} -> {ok, id} or a clear error (503 if RESEND_API_KEY unset, 400 bad input, 502 surfacing the Resend message). stdlib urllib, no new dependency. ACTION REQUIRED (Jeff): add RESEND_API_KEY and FROM_EMAIL=reports@affirmology.ai to the Render affirmology-studio-api service env, then redeploy. Wiring verified locally (route registered, correct 503/400/502 paths); the actual test send to jeff@jeffparker.love is PENDING the Resend key (Cowork has no key + no cloud studio token, so it cannot send from here). Verified live: /api/send-digest returns 401 (registered + gated).

  HERMES AVAILABILITY HOTFIX (SELF-HEAL + CLEAR) BUILT + DEPLOYED 2026-06-20 (commit 90dc89c; build tag 2026-06-20-hermes-selfheal-and-clear; app.js v24). Symptom: Hermes chat returned BadRequestError on EVERY message (even "hi"/"ye") for jeff on the CLOUD. DIAGNOSIS (evidence, not guess): the structural normalization (_normalize_history + most-recent-N list_hermes_history, commit fbf49e9) is already on origin/main and, replayed against the LIVE Anthropic API on jeff's real 20-turn history + a fresh "hi" + a 1.5M-char oversized turn, did NOT 400. A bare "hi" with the full system + all 12 tools also succeeded. So the live failure is the CLOUD running a build OLDER than fbf49e9 (its Render deploy never landed; the cloud build tag is unreadable from here because Cloudflare Access 302s /api/health). FIX (api/ + web/ only, no engine change, demo untouched): (1) SELF-HEAL in the hermes() loop: the first time Anthropic rejects the window, collapse to just the current user message (with any C22 doc context) and retry once, so a bad window can never again take the whole chat down; full exception logged. (2) CLEAR: jobs.clear_hermes_history + POST /api/hermes/clear + a "Clear chat" button in the Hermes header, the instant unblock if a thread wedges (user-initiated, scoped to the person). (3) BUILD_TAG bumped so the redeploy is verifiable at /api/health. Tests: tests/test_hermes_selfheal.py (self-heal on a forced 400 + clear wipes history), isolated against a tmp studio.db; 8/8 with the existing hermes-history tests. NOTE: could not clear jeff's CLOUD hermes_history from here (no cloud token + Access gate); the redeploy itself restores Hermes via self-heal, and the Clear button resets any thread.

  C22 ATTACH/PASTE FILES INTO HERMES CHAT BUILT + DEPLOYED 2026-06-20 (commit 7f7809c; app.js v25). User hands Hermes a .txt/.md/.pdf as READ-ONLY context for one question; never stored, never shared. Frontend (web/): paperclip attach button + hidden file input + paste-a-file on the textarea (intercepts only real files; text paste falls through); text read client-side, PDFs sent as base64; removable chips above the bar that ride the next message then clear; attachment-only send gets a default "read these" ask. Backend (api/main.py): HermesRequest.attachments [{filename, text?, pdf_base64?}]; PDFs extracted server-side with pypdf (already a dep, never raises, scanned/broken PDF degrades to a visible note); per-doc cap ~20k chars + total ~48k with truncation notes; labeled context ("Attached document :\n") prepended to the current user turn in the MODEL WINDOW ONLY, only the clean message persisted. Tests: tests/test_hermes_attachments.py (labeling, caps, PDF error path, and the proof that the doc reaches the model but is not stored); real PDF round-trip verified via a weasyprint-generated PDF; 9/9. C19 (stored natal basis) was already built+deployed earlier today (commit 8452cda); confirmed its acceptance tests still pass.

  HERMES SYNTHESIS-CREATE HARDENING BUILT + DEPLOYED 2026-06-20 (commit 0c2e2ed; build tag 2026-06-20-hermes-synthesis-selfheal). Symptom: Hermes surfaced BadRequestError again on a CAST (render_sacred_audio with deliver_to). ROOT CAUSE (evidence): all 12 HERMES_TOOLS schemas validate clean against Anthropic's rules (malformed-schema ruled out); the in-loop create() already self-heals (90dc89c), but the TWO forced-synthesis create() calls (the round-cap branch and the short/stub branch) were NOT wrapped, so a 400 on the accumulated tool-block window after a multi-round Cast propagated to the outer except and surfaced raw to the user. FIX (api/main.py only, no engine change, demo untouched): new _synth_create() wraps both synthesis calls (try full window with tools; on any error retry once on a flattened tools-free text window via new _flatten_messages_to_text(), which folds tool_result data into plain text and drops tool_use/tool_result blocks; returns None if even that fails, so the caller keeps the best text gathered so far). Regression test tests/test_hermes_cast_synthesis.py reproduces the exact Cast-with-deliver_to flow where synthesis 400s and asserts Hermes still returns a real reply (no raw BadRequestError); 24/24 studio tests green. Confirmed /api/health serves the new build tag (TestClient 200). NOTE: the currently-running LOCAL funnel server (port 8765) still reports the OLD build 2026-06-20-send-digest-resend, so it must be restarted (./start_public.sh) to pick up ALL of today's Hermes work (self-heal+clear 90dc89c, C22 7f7809c, synthesis hardening 0c2e2ed); the cloud picks them up on Render redeploy. The three live Hermes BadRequestError reports today were all consistent with the deployed server running code OLDER than the relevant fix.

  CLOUD RENDER PATH FIX (AGENT_PYTHON RESOLUTION + AUDIO PREFLIGHT) BUILT + DEPLOYED 2026-06-20 (commit cac707a; build tag 2026-06-20-cloud-render-agentpython-preflight). Symptom: a Cast on studio.affirmology.ai returned the script but no audio link/file; cloud /api/health showed agent_python_exists:false (agent_python "python"). DIAGNOSIS: both render phases shell out to subprocess.Popen([config.AGENT_PYTHON,"-m","affirmology.cli",...]) (api/pipeline.py run_script/run_audio). The cloud Dockerfile sets AGENT_PYTHON=python and pip-installs the agent into the image's main interpreter; config returned the bare "python" verbatim, so Path("python").exists() read False (a misleading health flag) and on any host where "python" is not on PATH the render subprocess would fail. The SCRIPT step needs only the Anthropic API; the AUDIO step also needs ELEVENLABS_API_KEY + the music bed, so the real audio-only blocker (most likely ELEVENLABS_API_KEY unset on Render, or the R2 music bed not present) showed up as "script yes, audio no" with no surfaced reason. FIX (api/ only, engine + demo untouched): (1) config._agent_python() now resolves to a REAL runnable interpreter (absolute existing override honored; bare override resolved on PATH via shutil.which; unresolvable override ignored -> local venv, else sys.executable, the dep-complete single-image interpreter), so agent_python_exists is true and the subprocess always runs. (2) pipeline.run_audio() preflight fails LOUD + CLEAR when ELEVENLABS_API_KEY is unset or the music bed is missing, surfaced to the chat via the existing _run_full_render failure delivery instead of dying silently. Tests: tests/test_cloud_render_path.py (override resolution incl. cloud bare-python, fallback, absolute-honored, health true, both audio preflights); 30/30 studio tests green; verified the resolved interpreter runs affirmology.cli --audio/--council; real local end-to-end Cast render verified (audio+PDF produced + delivered). ACTION REQUIRED (Jeff): on the Render affirmology-studio-api service, confirm ELEVENLABS_API_KEY is set and check cloud /api/health now shows agent_python_exists:true, elevenlabs_key_set:true, music_bed_exists:true; those three being true is the cloud-render green light. NOTE: a JobManager re-init (server restart, OR any concurrent import of api.main such as a pytest run against the live data/studio.db) reaps in-flight jobs as "Interrupted by a restart" via _reap_stale(); do not run the studio test suite against the live db while a real render is in flight.

  PERSONAL KNOWLEDGE BASE (GOLDEN-TIER CORE) BUILT + DEPLOYED 2026-06-20 (commit d028d81; build tag 2026-06-20-personal-knowledge-base). Per Affirmology_PersonalKnowledgeBase_GoldenTier_Concept_v1.md. The durable per-person standing-context store, the foundation of the golden tier and the real fix for context loss. (1) STORE: knowledge_base table in studio.db (id, person_id, title, content, kind [standing_context|reading|goal|note|suite], pinned, created_at, updated_at) with kb_add/get/list/list_all/update/delete/upsert in jobs.py; mirrored to R2 as ONE JSON object (api/kb_r2.py, STUDIO_KB_R2_KEY=knowledge_base/kb.json) on every write + restored on boot, so it survives a server REBUILD (not just a restart). No-op locally / if R2 unconfigured; never raises. (2) ALWAYS-INJECT: every Hermes turn appends the person's PINNED entries to the system prompt (main.py kb_standing_context), separate from + additional to chat history, so a chat clear or restart NEVER loses standing context; clear_hermes_history explicitly never touches the KB. (3) ADD/EDIT: Hermes tool save_to_knowledge_base {title, content, kind, pinned} (fires on "remember this / keep as standing context") + GET/POST/PATCH/DELETE /api/kb for the UI panel; circle-gated, each member manages their OWN entries (person_id scopes every write), outsiders (josh) rejected. (4) FEED AUDIO: create_studio_draft folds the person's pinned entries into the render brief (hermes_tools.pinned_kb_brief), so a Cast is bespoke to the accumulated context (ties to C24). (5) SEED: the standing WEALTH CONTEXT (from CLAUDE OUTPUTS/Affirmology/Affirmology_WealthCodes_JeffAndSol_v1.md) is the first pinned standing_context entry for Jeff and Sol, committed as text in api/seed_data/wealth{jeff,sol}.md and seeded on boot (idempotent, api/kb_seed.py, re-runnable python -m api.kb_seed), so both local + cloud seed identically. LIVE LOCAL DB SEEDED + VERIFIED (jeff 5.8k chars, sol 6.8k chars, both pinned; always-inject confirmed). Tests: tests/test_knowledge_base.py (persistence across a simulated restart, survival of a chat clear + always-inject, pinned-only injection, ownership scoping, idempotent seed, tool + permission, Cast brief, R2 no-op, endpoints); 39/39 studio tests green. api/ + render.yaml only, engine + demo untouched. NEXT (per concept doc): UI panel to view/edit the KB; grow toward the full Living Profile (C20); design the golden-tier credit model. ACTION (Jeff): restart the local funnel server to pick up the KB (and all prior Hermes/render fixes); the cloud auto-redeploys on push and boot-seeds itself.

  C24 CAST VOICES THE RIGHT SCRIPT BUILT + DEPLOYED 2026-06-20 (commit 327465c; build tag 2026-06-20-c24-cast-right-script). Benchmark: Affirmology_BeliefBreakthrough_SacredAudio_Jeff_v1.md. Problem: render_sacred_audio defaulted source_script to "from_chart", so a script Hermes composed in chat was ignored and the audio recited the bare natal blueprint. Fix (api/ only; the verbatim render-as-is plumbing landed earlier in 425dbda): (1) THE SCRIPT RULE, enforced in the tool description AND the Hermes system prompt: if Hermes wrote a script it MUST pass it as source_script (voiced VERBATIM); never Cast with an empty source_script when a script exists. (2) render_sacred_audio gains brief: when generating from the chart (no source_script), the conversation's theme/intent is woven into the render brief via _from_chart_brief() (a craft directive: themed, first-person, present-tense, SPEND the chart, do not recite placements) combined with the person's pinned knowledge base (pinned_kb_brief) and the live timing already in the council path; relationship branch carries the theme too. New schema fields document source_script + brief. Tests tests/test_c24_cast_brief.py (verbatim source_script rides through as the draft; a from-chart Cast for "wealth, belief, and breakthrough" puts theme + craft directive + standing context into the concept). LIVE VERIFY: a real from-chart council render for that theme produced a 5164-char themed first-person script that names the theme and reuses Jeff's standing anchor line ("Wealth flows to the order I keep") from his seeded Wealth Context, not a placement recitation; verbatim word-for-word was proven by the Morning Activation renders. 45/45 studio tests green. NOTE: C24 makes the from-chart path themed via the brief, but the actual first-person craft still depends on the council honoring the directive; the benchmark doc is the quality bar.

  RENDER OUTPUT NAMING + THEME/TYPE/AUTO-VERSION BUILT + DEPLOYED 2026-06-20 (commit 5302423; build tag 2026-06-20-render-naming-theme-type). New convention {FirstLast}{Theme}{Type}_v{N}.mp3 + {..}_v{N}_Report.pdf, applied in the STUDIO PIPELINE ONLY via the storage key after the integrity guard (engine + locked demo untouched; "Demo" kept only for the demo path). pipeline.render_basename(): FirstLast = first two name tokens (Jeff_Parker, Sol_Ballard); Theme = short CamelCase slug (default SacredAudio); Type = modality (default Custom, structure fallback via structures.structure_modality, before_bed->Bed); v{N} AUTO-INCREMENTS past any existing file with the same base. jobs: audio_theme/audio_type columns (migrated) + create_draft stores them + _run_audio keys the published mp3/pdf with the new base. hermes_tools.render_sacred_audio + create_studio_draft gain theme/type; the tool description + Hermes system prompt REQUIRE inferring + passing theme + type on every audio (version never asked). main: StudioDraftRequest gains theme/type. VERIFIED with two real renders: Jeff_Parker_WealthCodes_Morning_v1.mp3 + _v1_Report.pdf, then auto-named v2. Tests tests/test_render_naming.py.

  BREAKTHROUGH CRAFT AREA ADDED + SEEDED + SYNCED 2026-06-20 (commits d86378a + c122af5; build tag 2026-06-20-breakthrough-craft-area). Sixth craft area "breakthrough" (surfacing + rewriting limiting beliefs via reconsolidation, reframing, identity, rehearsal). Added to CRAFT_AREAS in BOTH craft.py copies (local engine + vendored) and to hermes_tools _CRAFT_AREAS (test test_craft_areas.py asserts the tool enum stays in sync with the engine areas); Affirmology_BreakthroughCraft_SeedSource_v1.md added to craft_seed SEED_DOCS (local engine; the vendored copy has no craft_seed.py, seeding is local-only). Seeder run (cheapest model, $0.21 of $1.50 cap): breakthrough=24 entries, craft total 94->108, brand-clean. R2 SYNC via merge-and-push: pulled R2 corpus/corpus.db, backed up to corpus/backups/corpus.db.pre-breakthrough-20260620, replaced ONLY the craft table with the local 108-row craft (integrity_check ok; meaning=20/distinctions=1 and record=25570 preserved), pushed back. FINAL craft by_area: audio_production 22, breakthrough 24, frequency_sound 6, meditation_hypnosis 15, neuroscience 10, reading_craft 31 (total 108). NOTE: re-running craft_seed re-distills ALL docs (non-deterministic) so existing areas shifted slightly. ALSO: craft_seed only authenticates if ANTHROPIC_API_KEY is in the env, so run it as: cd affirmology-agent && source .venv/bin/activate && set -a && source .env && set +a && python -m affirmology.corpus.craft_seed. ACTION (Jeff): restart the local funnel server (it has been on STALE build 2026-06-20-hermes-synthesis-selfheal for several tasks) to pick up the naming change, C24, KB, render fix, AND the breakthrough corpus.

  PRACTITIONERS & TECHNIQUES RESEARCH LANE ADDED TO NIGHTLY GROWTH 2026-06-20 (commit a21fbcf). Per Affirmology_PractitionersTechniques_ResearchCorpus_Plan_v1.md. New seventh craft area "practitioners" (the assess-later wisdom layer: teachers + techniques of transformation/manifestation). New module corpus/practitioners.py (sibling + vendored): a research_backlog table seeded from the canon (29 items: Wilber, Gurdjieff, Jung, Neville, Murphy, Coue, Erickson, NLP, Maltz, the living-from-the-end / wave-function family, etc.) and growable via add_to_backlog (e.g. Jeff's ChatGPT lists). research_lane() takes the next bounded batch, distills each via the cheapest model (injected for tests) into a KEPT craft entry: core method, mechanism (labeled evidenced/traditional/metaphor), the actionable practice, how it serves an audio, tags, usefulness/confidence score; copyright tier-wall (original method only, never verbatim; public-domain fuller; quantum/wave flagged as metaphor); no medical claims; no em dashes; NEVER auto-adopted (council reads it, a human/Apollo's gate graduates it). Wired in api/growth_cron.py AFTER the meaning stages with its own caps (GROWTH_PRACTITIONERS_MAX_COST_USD default 1.0, GROWTH_PRACTITIONERS_BATCH default 6) so it never starves the meaning builder; appends a morning-report section (teachers researched, entry counts, top new finds). CRAFT_AREAS (both) + hermes_tools _CRAFT_AREAS gain "practitioners" (kept in sync, tested). NOTE: "research" is LLM-distillation from the model's knowledge, like the existing growth stages (no live web fetch wired in the cron); a search backend can be added later. Runs on the next nightly cron (not run live here); tests/test_practitioners.py covers area+sync, backlog seed/grow, a distilled batch + advance + report, and the entry fields. 53/53 studio tests green. BACKLOG GROWN (commits e922b6e + 8ce25f8 + 612797c): canon 29 -> 57 -> 92 -> 110 (Jeff's ChatGPT archive: depth/somatic/manifestation, eastern wisdom/yogic, western esoteric/mystical, meaning/flow/growth psychology, trauma/mindfulness/breathwork, sound/frequency/journey, mind-body, indigenous/shamanic, stoic/taoist, performance, AND an 18-entry HYPNOTHERAPY block: Braid, McGill, Boyne, Yapko, Gilligan, Rossi, Kappas, Banyan, Marisa Peer/RTT + inductions/deepening/Milton Model/suggestion/regression/parts/post-hypnotic/self-hypnosis/suggestibility). SORTING FLAGS: each researched entry now also carries scope (now|future|archive) + form (audio|journaling|mindset|journey|breath|sound), validated and folded into tags as "scope:..., form:...,"; the report shows scope + form counts. RESEARCH PRIORITY: _pending orders the backlog by CATEGORY_PRIORITY (hypnotherapy then neuroscience_suggestion, the disciplines closest to our audios) before broader traditions. WIRED INTO HYPNOS: council.select_craft adds a dedicated PRACTITIONER METHODS slice (area="practitioners") and HYPNOS_SYSTEM lists practitioner methods + belief breakthrough as craft to draw on for induction/deepening/suggestion/post-hypnotic (both council.py copies synced; empty+skipped until the lane populates the area). Live backlog seeded to 110 pending (queue head is hypnotherapy). 55/55 studio tests green.

  API spend: started ~$29 of credit, ~$19 spent, roughly $9-10 left (confirm in Console at https://console.anthropic.com/settings/billing). Haiku 4.5 rates: $1/M in, $5/M out. Max plan does NOT cover API; credits are separate.

  ---

  4. Demo audio baseline (locked)

  Canonical spec: Affirmology_DemoAudio_Baseline_v1.md (with the underscore). Do not use Affirmology_DemoAudioBaseline_v1.md, that one is a superseded stub.

  Essentials: the Jeff Parker render is the reference (files in affirmology-agent/out/jeff_parker_demo_2026-05-21/). Voice ElevenLabs Charlotte, model eleven_turbo_v2_5, settings stability 0.65 / similarity 0.85 / style 0.0 / speaker boost on. Music bed Heavenly Circuit, voice-anchored mix, target -16 LUFS, -1 dBTP.

  SPEED LOCKED TO 1.0 (2026-06-15): the proven-clean gold-standard renders (Jeff v1/v2) were at speed 1.0; v3 was 1.05. The earlier auto-tune climbed to 1.12+ on longer scripts and caused pitch warble (Jeff reported the demo unlistenable). The demo server now passes --speed 1.0 explicitly (locked), and the CLI auto-tune ceiling is capped at 1.05 so even the auto path can never warble. THE DEMO IS LOCKED: do not change its structure or settings without Jeff's explicit say-so. All experiments live in the Studio.

  QC NOW WIRED (2026-06-15): the failure mode is ElevenLabs Turbo dropping voice mid-track at high speed (Sol's/Alex's old renders). scripts/audio_qc.py runs on the voice-only track of every CLI --audio render, re-renders slower on a dropout, and refuses to serve a bad file. Note: QC catches dead-air dropouts, not pitch warble; the speed-1.0 lock is what prevents the warble. PDF: WeasyPrint was never installed (why demo PDFs failed); now installed, and on macOS the pipeline injects DYLD_FALLBACK_LIBRARY_PATH=/opt/homebrew/lib so WeasyPrint finds pango/glib. PDF confirmed generating.

  ---

  5. Demo backend (laptop, this week)

  Architecture: demo.affirmology.ai (Cloudflare Pages) -> Cloudflare Worker -> laptop over Tailscale -> FastAPI (affirmology-demo-site/server/api.py) -> affirmology CLI -> audio + PDF -> webhook -> Resend email.

  Done: Tailscale installed and connected on the laptop. Machine name macbook-pro-2, Tailscale IP 100.98.80.60. That IP is the Worker's MAC_MINI_URL target.

  Front-end (rebuilt 2026-06-13, in CLAUDE OUTPUTS/Affirmology/demo-site-v2/): cinematic landing with a short NDA gate and the intake form (index.html), and the result experience (result.html): a ~2 minute web-native teaser (kinetic text in the brand palette, scored by the Lily trailer VO plus music bed at assets/teaser.mp3), then a ritual cue, the personalized audio player, an educational outro, a partial report reveal, and download buttons for the MP3 and the PDF. Hero line is now "One voice. One chart. One healing." The form posts to /api/render; the result page polls /api/status/{job_id}. No bundled example is loaded. A clean Cloudflare Pages upload folder (index.html, result.html, assets/teaser.mp3 only) is at CLAUDE OUTPUTS/Affirmology/deploy/cloudflare-pages-upload/. The "video while audio loads" slot is web-native animation, not Higgsfield; a separate chat is also building an animation variant for it. Deploying this folder to the existing direct-upload Pages project (demo.affirmology.ai) via New deployment is in progress. Until the Worker and Tailscale are wired, the page holds after the teaser (no audio) by design. Future automation option: connect a GitHub repo to the Pages project for auto-deploy instead of manual upload.

  Single upload folder (2026-06-13): UPLOAD-THIS-TO-CLOUDFLARE/ at the AFFIRMOLOGY root is the one place to deploy. It now holds index.html (NDA forced every visit), result.html, assets/teaser.mp3, and the two new self-contained animation pages copied from the marketing chat: primer.html (= Affirmology_Primer_v5, the loading film with the lit-up bodygraph, Gene Keys wheel, audio-wave, Lily VO) and whatsinstore.html (= Affirmology_WhatsInStore_v2, the upsell reel). Uploading this folder makes all of them live (demo.affirmology.ai/primer.html, /whatsinstore.html). FUSED (2026-06-13): the flow is now form -> primer.html#<job_id> (the cosmic film with Lily VO) -> when the film ends it hands the job to result.html#<job_id>&film, which skips its own teaser, waits for the render (holding on "Composing your soul song"), then ritual ("Play my audio") -> audio player -> educational outro -> partial report reveal -> the two download buttons wired to the rendered files from /api/status. The in-page kinetic teaser stays as the fallback for a direct visit to result.html with no job. Standalone /primer.html (no job in hash) still shows its own endcard. Edits: index.html go() points to primer.html; primer toEnd() redirects when a job_id is present in the hash; result.html parses &film and runs film-mode. Also fixed a null name-wait reference that could have broken result.html. NEEDS Jeff's on-laptop live test (submit the form while the server and tailscale funnel 8443 are running). Re-upload the UPLOAD-THIS-TO-CLOUDFLARE folder to pick up the fused index.html, result.html, primer.html. POST-AUDIO FUSED (2026-06-14): when the personal audio ends, result.html now hands off to whatsinstore.html#<job_id> (the use-case reel). The reel's endcard wires its two buttons (Download your audio, Download your report) to the real files from /api/status and renders a partial report blueprint BELOW the buttons (buttons above the writeup, by request). So the re-upload now also carries the updated result.html and whatsinstore.html.

  DEPLOY ARCHITECTURE CHANGED (2026-06-15): the site is now a GitHub-connected Cloudflare Worker named affirmology-site (GitHub repo jeffparkerlove22/affirmology-site, local working copy at AFFIRMOLOGY/affirmology-site/), and it AUTO-DEPLOYS on git push. It serves demo.affirmology.ai (the custom domain was moved off the old direct-upload Pages project affirmology-demo, which is now orphaned and can be deleted later). The API worker affirmology-demo-worker still owns demo.affirmology.ai/api/* and was verified still working alongside the new worker. Canonical site files now live in AFFIRMOLOGY/affirmology-site/ (index.html, result.html, primer.html, whatsinstore.html, assets/teaser.mp3); Cloudflare also added its own Workers config commit to the repo. DEPLOY WORKFLOW going forward: edit the affirmology-site files, then in GitHub Desktop Pull (to pick up Cloudflare's config commit), Commit, Push; it auto-deploys in about a minute. There is also a test URL (affirmology-site.jeff-e97.workers.dev) for checking before/without touching the domain. The old UPLOAD-THIS-TO-CLOUDFLARE drag-and-drop path is retired.

  VERIFIED END TO END (2026-06-15): a real form submission on demo.affirmology.ai ran the full flow, form to primer film to personalized audio (rendered on the laptop, served over the Tailscale funnel) to the post-audio reel to the report and download buttons. Server terminal confirmed POST /render-sacred-audio 200, GET /status/{id} 200, and GET /files/.../.mp3 200/206. So /api/ on demo.affirmology.ai still routes correctly to the API worker even though the new static worker holds the custom domain (the earlier empty external probes were caching, not a routing break). The corpus-grounded, style-only generator was the version used. Open improvement notes from Jeff (to capture and work later): the audio, the report, the "video" pieces, and the overall process all have refinements he wants.

  LIVE (2026-06-13): FastAPI server (affirmology-demo-site/server/api.py) running on the laptop at port 8443, exposed publicly via Tailscale Funnel at https://macbook-pro-2.tail333640.ts.net . Worker deployed with MAC_MINI_URL and MAC_MINI_AUTH_TOKEN; shared secret is affirmology-demo-secret-2026; route demo.affirmology.ai/api/* active. Chain proven externally: /api/health and /api/status/{id} both answer through the Worker from the laptop. Server adds GET /status/{job_id} (returns state + audio_url + pdf_url + blueprint) and serves the rendered files at /files via the Funnel; Worker adds GET /api/status/{job_id} proxy. Resend domain affirmology.ai is verified, but RESEND_API_KEY and FROM_EMAIL are not yet added to the Worker, so email is still off (optional). First real form submission was fired 2026-06-13 for Jeffrey Alan Parker (job 535b38a3): the chain worked, the laptop accepted it and began rendering, but it was still rendering after ~7 minutes (slower than the usual 3-5, likely the concurrent corpus runs competing for CPU/API) when the laptop had to go offline for an investor meeting, so this render did not finish. Not a failure of the wiring; the POST and status polling both worked.

  CRITICAL operational fact: the live custom-audio demo only works while THIS laptop is awake, online, and running both the server and tailscale funnel 8443. When the laptop sleeps or leaves internet, the demo's audio generation is offline (Sol cannot test it remotely for this reason). The Cloudflare side (the site, the teaser, primer.html, whatsinstore.html) stays live regardless; only the per-person render needs the laptop.

  Restart recipe (Worker + secrets persist, no redeploy; funnel host stays macbook-pro-2.tail333640.ts.net): 1. Server window: cd affirmology-agent && source .venv/bin/activate && cd ../affirmology-demo-site then run uvicorn with AFFIRMOLOGY_AUTH_TOKEN="affirmology-demo-secret-2026", AFFIRMOLOGY_PUBLIC_BASE_URL="https://macbook-pro-2.tail333640.ts.net", PYTHONPATH=".../affirmology-agent/src", uvicorn server.api:app --host 127.0.0.1 --port 8443. 2. Tunnel window: /Applications/Tailscale.app/Contents/MacOS/Tailscale funnel 8443. Full commands in CLAUDE OUTPUTS/Affirmology/deploy/BACKEND-TEST-RUNBOOK.md.

  Still to do: complete one real render end to end (re-run a submission once the laptop is back online with the server up); consider whether the background corpus runs should pause during a live demo so renders finish in 3-5 min; add RESEND_API_KEY + FROM_EMAIL to the Worker to turn email on; re-upload already done for the always-on NDA + primer.html + whatsinstore.html.

  Hard rules for the demo: - The live demo must ALWAYS generate the custom audio for the exact person and chart entered in the form. Never load a bundled or stock example. The form posts to /api/render and the result page polls /api/status/{job_id}. - Do not use the Alex Carper render as an example anywhere; it has a mid-track audio volume problem (the dropout failure mode). - Cloudflare dashboard navigation (Jeff's account): Pages and Workers live under the sidebar item Compute (Workers), not "Workers & Pages." Inside a website (zone) the sidebar only shows zone tools, so click the Cloudflare logo to return to Account Home first.

  ---

  6. Brand and writing rules (enforce in all output)

  • STYLE IS SHARED, CONTENT IS NEVER SHARED. This is the core premise of the product. Reference scripts and "this one clicked" examples are reused only for craft: poetry, cadence, structure, proportions, length. A person's audio content (placements, signs, gates, gifts, numbers, images, identity titles) must come ONLY from their own chart and their own corpus sources. No founder's or any other person's chart data may ever appear in someone else's audio. This is baked into script_generator.py as rule 0, and the full recipe plus a fictional exemplar live in Affirmology_ScriptStyle_Reference_v1.md.
  • EVERYTHING IS CHART-DRIVEN. Every audio is generated from a specific person's unique chart, always. We build plug-and-play STRUCTURES (templates) that customize to anyone's chart. A "concept" is the high-level design of an audio TYPE (which systems lead, the focus, tone, length), never a chartless finished script. Scalable by design. Studio spec: Affirmology_Studio_Spec_v1.md.
  • Never use em dashes. Anywhere. Use periods, commas, or rewrite.
  • "Subconscious Operating System" is the core brand frame.
  • Sol = Soledad Gabriela Ballard. Never "Soul" for her name.
  • Joe Dispenza: OK to name as an EXTERNAL REFERENCE POINT (what is already out there, to differentiate and show the trend is mainstreaming). Do NOT describe Affirmology's OWN method as his, do not imply affiliation, and do not use his trademarked program names. For our own heart-coherence opening, call it "heart coherence breath" or "embodied gratitude induction." (Updated 2026-06-13 per Jeff: the earlier blanket "never name him" rule was too strict; the real intent is to not position ourselves AS him.)
  • Tier C copyrighted material (Richard Rudd Gene Keys official, Ra Uru Hu / Jovian Archive) is named for benchmark only and NEVER scraped. Retrieval helpers are tier-walled to A/B.
  • Do not practice therapy or give it without licensing.

  ---

  7. People

  • Jeff Parker Love (co-founder, equal billing with Sol; do NOT frame him as "the system" or label him "founder" in a way that slights Sol, he is more than a system): leads the Miami community. Born 1982-09-11 16:19 Anderson SC. Virgo Sun, Cancer Moon, Capricorn Rising, Life Path 22. Birthday September 11. Human Design (confirmed from his official report 2026-06-15): Manifesting Generator, Sacral authority, 2/5 Hermit/Heretic profile, Single Definition, Right Angle Cross of Rulership 3. FIXED 2026-06-15: human_design.py _derive_type now detects a motor-to-throat connection through the full defined-centers graph (indirect connections through the Spleen/G/etc. count, not just a single direct channel), so the engine returns Manifesting Generator for Jeff. His chart was otherwise correct (profile 2/5, Single definition, Sacral authority all matched his report, so birth data and gate calc are fine). Verified on his real channels (16/48, 26/44, 34/57) and control cases; Josh's oracle test is unaffected because his Throat is undefined. Re-run pytest on the Mac and re-render to confirm in the full pipeline (the server spawns a fresh process per render, so no restart is needed).
  • Sol Ballard (Soledad Gabriela Ballard): co-founder, relational and brand lead. (NOTE 2026-06-17, Jeff: the Miami community is JEFF's, not Sol's. Do not attribute the Miami community to Sol.) Born 1986-01-23 10:28 El Paso TX. Aquarius Sun, Cancer Moon, Aries Rising, Generator 2/4, Emotional authority, Life Path 3. Striking business synastry with Jeff (exact Cancer Moon conjunction, reversed nodal mirror).
  • Colin Jacobs: advisor (investor strategy, Norm Adams insight). Born 1976-04-17 14:30 Boston MA. Aries Sun, Virgo Rising, Birthday 8.
  • Josh Parini: early demo subject, strong reading expected, to be folded into investor materials. Born 1985-10-21 06:00 San José, Costa Rica. Generator 2/4, Sacral.
  • Norm Adams: prospective investor, shares Jeff's September 11 birthday, invests to belong to a movement.
  • ALL FOUR core charts (Jeff, Sol, Colin, Josh) are confirmed and live in affirmology-agent/src/affirmology/verification/known_charts.py and the May 29 renders in affirmology-agent/out/. They seed the Studio People Library so birth data is never re-typed.

  ---

  8. Key documents (what to open for what)

  • This file: current state, always first.
  • Affirmology_DemoAudio_Baseline_v1.md: locked audio spec.
  • Affirmology_ScriptStyle_Reference_v1.md: the script craft recipe and a fictional style exemplar (style shared, content never).
  • Affirmology_CloudStudio_BuildSpec_v1.md: AUTHORITATIVE build plan for moving the studio to the cloud (Render API + Cloudflare Pages + R2, phone-usable, 24/7, no laptop). MVP target before a Tennessee trip ~2 weeks out. Being built in a Claude Code session. Supersedes laptop+Tailscale as the target; the laptop demo runs interim. The studio FEATURE design lives in Affirmology_Studio_Spec_v1.md and layers on top.
  • Affirmology_InvestorVideoScriptBrief_v2.md: source brief for the investor video script and investor details (voice direction, differentiation, June 20 launch astrology, Jeff/Sol chart, macro moment, Loft, CTA).
  • Affirmology_SubconsciousOS_AgenticArchitecture_v1.md: the agent architecture (three tiers, corpus-fed generation design).
  • Affirmology_BackendArchitecture_v1.md: what is built across the five layers.
  • Affirmology_AudioMastering_Production_Research_v1.md: the mastering craft and tool research.
  • Affirmology_AudioSetup_Checklist_v1.md: install list, free audio tools, LANDR.
  • Affirmology_CorpusStatus_2026-06-13_v1.md: detailed corpus status snapshot.
  • Affirmology_MacMiniSetup_Guide_v1.md: Mac mini install path (paused).
  • Affirmology_DemoLaunchPlan_v2.md: canonical demo launch plan.
  • affirmology-agent/: the code. affirmology-demo-site/: the web demo scaffold.

  Marketing / video assets (built 2026-06-13)

  • Affirmology_Primer_v5.html (LATEST; v1-v4 are superseded iterations): the consumer "loading trailer" web experience that plays while the demo audio renders. Single self-contained file (Lily VO embedded), zero external dependencies (built-in Web Animations + CSS; do NOT reintroduce a CDN animation library, it broke in preview). Cosmic scenes: chart forms, then sustained Astrology / Gene Keys / Human Design showcases (with the centers and channels lighting up) with big titles and concept word-throw, the turn, a warm audio-wave bridge (concentric sound rings + stacked waveforms, no figure), bloom. Highlights-only big text, gentle typewriter, timing frame-accurate, non-overlapping word slots. It now ends on a "Your soul song is ready" hand-off with TWO inert download boxes (Download your audio, Download your report PDF), to be wired to the rendered files later. No membership/preparing language.
  • Affirmology_WhatsInStore_v2.html (LATEST): short post-audio upsell/concept film, now with a Lily VO (Affirmology_WhatsInStore_VO_Lily_v1.mp3, ~65s) hammering use cases (breathwork/journey priming, morning walk, run, gym, pre-sleep, relationships, abundance, healing, the transit cadence, the chatbot guide), big held use-case lines, an audio-wave/equalizer motif, ending on the SAME two inert download boxes (Download your audio, Download your report PDF). NOTE: concept/marketing piece, not the real results screen. The real results page (user's astrology text + the two real download links) is built in the separate webpage chat; see the handoff doc.
  • TRAILER VOICE DECISION (distinct from demo + investor voices): the consumer trailer uses ElevenLabs Lily (velvety British, voice id pFZP5JQG7iQjIQuC4Bku), generated at speed 0.7 then time-stretched atempo 0.95, synced via the with-timestamps endpoint. File: Affirmology_TrailerVO_Lily_synced_v1.mp3 (~109s). The trailer script is Sol's edit (in Affirmology_TrailerVO_Sol_v1_script.md). Note: demo-audio voice is still Charlotte (section 4); investor-video narrator should be the male voice per the v2 brief. Three different voices for three different assets, on purpose.
  • Strategy/source docs: Affirmology_VideoStrategy_v3.md (trailer + investor film strategy, evidence vault), Affirmology_TrailerProductionKit_v1.md, Affirmology_InvestorVideoKit_v1.md, Affirmology_InvestorBrief_v1.pdf/.html (written investor brief, terms section left open for Jeff), Affirmology_PredictiveBrain_IdentityChange_Research_v1.md, Affirmology_CreatorEconomy_Monetization_Research_v1.md, Affirmology_WebFlow_Handoff_v1.md (copy-paste handoff for the webpage-building chat: the consumer flow, the two download buttons, design tokens, brand rules, voices).
  • ElevenLabs API key scope (this key): text_to_speech and with-timestamps WORK; music_generation and speech_to_text are NOT permitted (401 missing_permissions). Music beds come from Suno (prompt in the Trailer Kit); add under the trailer later.
  • INVESTOR VIDEO RECONCILIATION (2026-06-13): Affirmology_InvestorVideoKit_v1.md was rewritten to follow Affirmology_InvestorVideoScriptBrief_v2.md (the governing brief): male narrator, Subconscious Operating System frame, moat reframed honestly as taste + Sol/Miami community + Jeff + velocity (NOT the database), and the June 20 launch chart, Jeff-and-Sol synastry, Aphrodite name layer, macro "AI reshaping work" moment, the Affirmology Loft, the Affirmologist/affiliate ceremony network, belong-to-the-movement investor psychology, and multi-door CTA all added. Narrator pick still open (test George/Brian/Bill). STILL TO DO: the written Affirmology_InvestorBrief_v1 PDF and the investor section of VideoStrategy_v3.md predate the v2 brief and still lead with the database/bridge moat and older rollout; bring them to the brief's framing in a future pass.
  • INVESTOR/MISSION FILM CONCEPT + TOOLING (2026-06-14, captured in Affirmology_InvestorVideoKit_v1.md Parts G and H, NOT yet scripted): reframed as a participation/mission film (recruiting beta testers, ambassadors/affiliates, and investors/team, not just investors). Two host characters in dialogue, both as LIVING MARBLE-AND-GOLD STATUES with inner constellations: a robed Hellenistic Oracle (Urania/Sophia wisdom, intact and draped, NOT a nude Venus) and a timeless Astronomer-Sage (philosopher-astronomer, NOT a fantasy wizard). They mirror Sol and Jeff; Jeff and Sol appear in their own office-studio clips near the end. Includes an objection-handling map, a 'how it works' agentic-system/backend visualization beat, and the ask (tiers $5k-$30k, target $30k+ this summer, $50k+ desired, small amounts welcome; numbers stay off screen). CAPITAL PHILOSOPHY (2026-06-15, Jeff, important for all investor framing): NOT raising millions and NOT courting VC firms. Total target roughly $30k to $150k, up to ~$250k, to break through and stand up the Loft/engine, after which the business is profitable via ads, affiliates, events, and product revenue. Direct Wyoming LLC units are REAL OWNERSHIP (not SAFEs): the right to share in profit exists from the start, BUT actual distributions are realistically 12-24 months out, after break-through and profitability (Jeff was explicit 2026-06-15: do NOT claim "distributions from day one" or near-term cash; be honest about timing). The "steal" rests on a low entry valuation on a live, working product before revenue reprices it, not on immediate income (Jeff: "people seeding with us are getting a steal"). The entity stays an LLC on purpose (the C-corp/VC path is explicitly NOT the plan). New investor docs (2026-06-15): Affirmology_NorthStar_v1.md (first-person founder vision/conviction narrative, the data-room centerpiece), Affirmology_InvestorDataRoom_Blueprint_v1.md (data-room gap analysis + build order inspired by the YouPoff pre-seed breakdown). TOOLING decided: GSAP is now 100% free (use on the live in-browser web pages); Higgsfield generates the cinematic footage. ON-SCREEN GRAPHICS TOOLING (updated 2026-06-15, Jeff): build all motion graphics in BOTH Remotion AND Hyperframes IN PARALLEL and A/B which is better per need until a clear winner emerges (if ever); applies to the V2 website videos (teaser + post audio) AND the investor film. Both render via headless Chrome + FFmpeg on Jeff's Mac and both export transparent ProRes 4444 overlays. Remotion = React (free <=3 employees, mature, great for high-volume-from-data); Hyperframes = HeyGen's open-source HTML+GSAP, agent-native (Claude authors compositions in plain HTML, our strongest format), likely the lower-friction fit for how we work. Remotion starter already scaffolded in Affirmology_RemotionGraphics/. When full per-video plans are written, start creating the actual assets here (Claude authors, Jeff renders). Mac render prereqs: Node 22+ and FFmpeg. Full map: Affirmology_VideoToolingPlan_v1.md; research: Affirmology_VideoAsCode_Research_v1.md; friends-and-family raise structure: Affirmology_FriendsAndFamilyRound_v2.pdf (v2 SUPERSEDES v1; v1 wrongly implied day-one distributions, do not send it). North Star narrative: Affirmology_NorthStar_v1.md.

  ---

  9. Open threads / next actions (prioritized)

  Full demo feedback from the 2026-06-15 run-through is tracked in Affirmology_DemoPunchList_v1.md. Resolved 2026-06-15: HD type (Manifesting Generator fix held), the audio (speed-1.0 lock + QC gate, voice-anchored 6:17 mix covers the voice), and PDF generation (WeasyPrint installed + lib path). Still open: email needs the Resend key added to the worker; the primer film needs gap-filling text (all sphere names, more astrology points) and subconscious-block visuals instead of Human Design on the "bypass" beat; compare on-screen blueprint vs PDF vs original baseline for consistency.

  1. DONE 2026-06-15: demo audio warble FIXED (locked to speed 1.0, auto-tune ceiling 1.05), audio_qc.py WIRED into every render, WeasyPrint installed so the demo PDF generates. Re-rendered Jeff clean (QC PASS, 6:17, PDF OK). The demo is LOCKED.
  2. DONE 2026-06-15: Cloud Studio built and verified end to end (affirmology-studio/, section 11). Local-first, two-phase script-then-audio, named structures, QC-gated audio, mobile-first brand-matched page. Run with affirmology-studio/run_local.sh.
  3. NEXT for the Studio: layer the Studio Spec features (comfort vote 1 to 5, A/B variants, multi-person feedback, leaderboard, push-to-inner-circle, Google Sheet tracking, known-chart accuracy harness); tune per-structure length (a couple overran the 950 word cap); structure the neuroscience research into a "techniques" corpus tradition the generator pulls from.
  4. Cloud deploy the Studio when Jeff has accounts: Render (Docker from render.yaml) + R2 bucket (STUDIO_STORAGE=r2) + Cloudflare Pages for web/ + Cloudflare Access gating on studio.affirmology.ai. All scaffolding is in the repo.
  5. DONE 2026-06-14: corpus wired into generation via _corpus_sources() in script_generator.py; v5 reference is a content-free skeleton. Refine retrieval keywords if snippets read noisy.
  6. Finish the human design structuring backlog (in budget). Then, with more credit, re-mine the big western/vedic/numerology books deeper.
  7. Demo backend hardening: add RESEND_API_KEY + FROM_EMAIL to the worker to turn email on (still off). The laptop must be awake + server + funnel running for the live custom render.
  8. Build the nightly launchd watchdog so the corpus and QC run and report automatically.
  9. Investor video: fold in Josh Parini's reading and any Colin material when available; pick the male narrator.
  10. Mac mini: resume when the USB hub is replaced.

  ---

  10. Traps and stale-info warnings

  • Do not trust a tradition's "structured %" alone; check the word_count floor (sub-100-word docs are skipped by design).
  • The corpus now lives on the INTERNAL disk at /Users/jeffreyparker/CLAUDE/AFFIRMOLOGY/corpus (moved off the SSD 2026-06-18, section 20). Renders/studio-media still persist to the SSD (/Volumes/Affirmology/studio-media). A cloud sandbox cannot see local files, and Terminal is locked to click-only under screen control, so credit-spending commands run on the Mac and are pasted by hand. This is why the human runs commands, not the agent.
  • Activate the venv (or call .venv/bin/python) before running the agent; a fresh reboot drops the venv.
  • Run audio QC on the voice-only track, not the final mix (music masks dropouts).
  • Do not hand-edit .claude/settings.local.json: the permission UI owns it and rewrites it on every grant, clobbering edits. Put durable allow rules in .claude/settings.json (project), which merges and is not auto-overwritten. A broad safe allowlist (python, curl, grep, pip, uvicorn, pytest, file edits) plus defaultMode: acceptEdits is already there to reduce permission prompts during infra work.

  ---

  11. Cloud Studio (built 2026-06-15, runs local-first)

  Governing build spec: Affirmology_CloudStudio_BuildSpec_v1.md (infrastructure). Feature design: Affirmology_Studio_Spec_v1.md. Both honored. The app is affirmology-studio/ at the AFFIRMOLOGY root.

  What it is: one app that is BOTH Jeff's desktop tool and the future studio.affirmology.ai. It wraps the existing affirmology-agent pipeline (does not rewrite it). Cloud-shaped (FastAPI api/, static web/, storage.py abstraction, api/Dockerfile, render.yaml) but runs LOCAL-FIRST on this laptop using the agent's venv and .env. Storage is local now (files served at /media), R2 is a drop-in swap (STUDIO_STORAGE=r2 + R2_* env, R2Storage already written).

  Two-phase, script-first cost discipline (matches the hard rule): - "Generate script" (cheap default): CLI --script --build-report --concept <structure directive>. Strong model for the script, Haiku for the script map and build report. No voice cost. - "Render the audio" (on demand, keepers only): CLI --audio --music <bed> --speed 1.0, QC-gated, into the same job folder so it reuses the script. Produces the mixed MP3 + the Sacred Audio Report PDF.

  Structures are DEFINED, NAMED templates: api/structures.py has 14 structures across 8 categories (Foundational, Gene Keys, Astrology, Human Design, Numerology, Subconscious & somatic, Integrative, Custom). Each locks a shape (which systems lead, sequence, technique slots, length, tone) as a directive; the chart supplies all content; the model never re-invents the shape. "Custom" takes a free directive.

  Endpoints (api/main.py): GET /api/structures, POST /api/script, POST /api/audio/{job_id}, GET /api/jobs/{id}, GET /api/jobs (the library), GET /api/health. GATE: local dev runs WITH NO GATE (2026-06-15). ROOT CAUSE of the "stuck on the access screen / button does nothing" bug was CSS, not auth: .gate sets display:grid, and an author display rule overrides the browser default [hidden]{display:none}, so toggling the hidden attribute in JS never actually hid the gate (it stayed painted as a fixed overlay even though the code had dismissed it and loaded the shell underneath). Fixed with [hidden]{display:none !important;} in styles.css. Separately, the gate is now also removed for local entirely: config.REQUIRE_AUTH (env STUDIO_REQUIRE_AUTH, default false) controls it: when false the API does not require the token and the front-end skips the gate entirely (it reads health.config.require_auth and lands straight on the compose screen). The cloud deploy sets STUDIO_REQUIRE_AUTH=true (in render.yaml) so studio.affirmology.ai stays gated by the shared bearer token (STUDIO_AUTH_TOKEN, default affirmology-studio-local-dev). Front-end asset refs are cache-busted (?v=2). Jobs run one-at-a-time in a thread, state in SQLite (affirmology-studio/data/studio.db), so the library survives a restart. Web (web/index.html|styles.css|app.js) is mobile-first and brand-matched to the live site (forest green / gold / cream; Cormorant Garamond + Inter + JetBrains Mono): gate, compose (person + structure), live progress, script view (script + build report + chart chips + "Render the audio"), result view (player + MP3/PDF downloads), and the library.

  VERIFIED END TO END (2026-06-15): script phase produced a grounded 1009-word script (0 em dashes, correct chart: Manifesting Generator) + build report on Haiku; audio phase rendered Charlotte at 1.0, QC PASS, mixed MP3 + PDF, both served via /media. 29 pytest green after all engine changes.

  FRONT-END REBUILT TO THE APPROVED MOCKUP (2026-06-15, later same day): per Affirmology_StudioBuildBrief_v1.md, the front-end now matches CLAUDE OUTPUTS/Affirmology/Affirmology_StudioUI_v3.html exactly (locked brand: starfield + aura + grain, Cormorant/Inter/JetBrains, emerald+gold). It is a 5-screen SPA wired to the real engine: LOGIN (local: log in as any of the four seeded people), HOME (greeting, quick actions, People Library with REAL computed charts, activity feed from real jobs, approved picks, category grid with readiness badges ✓ working / empty / in progress / ★ pick), COMPOSE (conversational workbench: paste-anything brief -> real chart-driven script via the two-phase engine; sending more messages makes new versions; gold "Render to audio" spends voice credits only on keepers; right rail with person/category select/length/status/versions), LIBRARY (per category, empty states + working-version badge across people), VERSION DETAIL (player, full script, "Why this build" = the team build report + Audio QC status, feedback thread with comfort vote 1-5 + comments, Make-this-the-pick). NEW BACKEND: api/people.py seeds four charts (Jeff, Sol, Colin, Josh) from known_charts.py, which now also holds SOL_BIRTH and COLIN_BIRTH. Endpoints: /api/people, /api/categories, /api/categories/{id}, /api/compose, /api/audio/{id}, /api/jobs[/{id}], /api/jobs/{id}/feedback, /api/jobs/{id}/pick. Jobs are organized as PERSON + CATEGORY + VERSION with picks and feedback in SQLite. Deep-link #as=<person>[&pg=<page>][&job=<id>] logs straight in. VERIFIED (2026-06-15): pytest 29 green; real two-phase render for Sol / Before-bed reprogramming (826-word script, 0 em dashes; Audio QC PASS; MP3 9.9MB + PDF 71KB served via /media; build report 11.6k chars); feedback + pick persist; category readiness updates; home screen screenshotted and matches the mockup, with the live render showing in the activity feed. FAST-FOLLOW (UI present, not yet wired, per the brief): relationship/multi-chart synergy, deep cosmic report (PDF, no audio), new-category creation, share/required-request notifications. The Verifier is not yet surfaced in the Studio build report (Audio QC is).

  Run it locally: cd affirmology-studio && ./run_local.sh then open http://localhost:8000. Access key not needed locally (gate-free on localhost).

  PHASE 1 DEPLOY = LAPTOP ENGINE + CLOUDFLARE (2026-06-15, per Affirmology_StudioDeploy_Runbook_v1.md; NOT Render yet, that is Phase 2 this week). The WHOLE Studio (UI, API, script phase, audio renders, SQLite library) runs on the laptop. PERSISTENCE: finished media (MP3 + PDF + build report) persist to the SSD at /Volumes/Affirmology/studio-media (env STUDIO_MEDIA_DIR); the library DB stays on the laptop at affirmology-studio/data/studio.db; render scratch in data/renders/. Both survive restarts (verified). START (two terminals): cd affirmology-studio && ./start_public.sh (server on 127.0.0.1:8765, prints the token; sets STUDIO_REQUIRE_AUTH=true, media on SSD) and /Applications/Tailscale.app/Contents/MacOS/Tailscale funnel --bg --https=8443 8765. Public funnel URL: https://macbook-pro-2.tail333640.ts.net:8443 (the demo's funnel on 443 is untouched; studio uses 8443). AUTH: the token is required for funnel/public traffic (the Cloudflare worker injects it); direct http://localhost use on the laptop is exempt by Host header, so the laptop tool stays gate-free. Token persisted in affirmology-studio/.studio_token (gitignored). CLOUDFLARE (Jeff clicks, click-by-click in affirmology-studio/PHASE1_CHECKLIST.md): front-end on Pages (Direct Upload of web/, no GitHub needed) at studio.affirmology.ai; a Worker (affirmology-studio/cloudflare-worker/) routes studio.affirmology.ai/api/* AND /media/* to the funnel URL with the token injected (secrets STUDIO_FUNNEL_URL, STUDIO_AUTH_TOKEN); Cloudflare Access (Zero Trust) one-time-PIN allowlist for jeff@jeffparker.love and solballard@gmail.com ONLY. VERIFIED 2026-06-15 (laptop + funnel side; the studio.affirmology.ai hop is Jeff's clicks): pytest 29 green; whole app runs on the laptop; media persists to SSD, DB on laptop; a rendered audio + the library survive a server restart; script phase works through the public funnel (Jeff/Founder confidence, 898 words, 0 em dashes); the ONE approved audio render persisted to the SSD and plays back over the funnel (HTTP 206, seekable); token enforced over the funnel (401 without it), localhost gate-free. UI fixes done: logo is Affirm + gold ology; Jeff and Sol both labeled "Founder" (equal billing). COST DISCIPLINE: script-only by default; audio (ElevenLabs) only on explicit per-version approval; the engine never auto-renders. KEEP READY for Phase 2 (a config switch, not a rebuild): render.yaml, api/Dockerfile, the R2 backend in api/storage.py, and STUDIO_REQUIRE_AUTH all in place.

  STUDIO UI APPROVED 2026-06-15 (Jeff signed off on the direction): the clickable mockup CLAUDE OUTPUTS/Affirmology/Affirmology_StudioUI_v3.html is now the DESIGN SOURCE OF TRUTH, and Affirmology_StudioBuildBrief_v1.md is the authoritative build brief for Claude Code. The approved Studio is more than the current build: (1) Compose is a CONVERSATIONAL workbench, one big paste-anything box (no char limit, takes a whole Gemini dump), the system reflects back / asks questions / drafts a script inline / revises as you type, until you say "I want to hear this version" then the render button spends voice credits (script-first cost discipline preserved). (2) The logged-in person is the DEFAULT subject; you can compose for someone else and both can see/hear it (testing). (3) Per-category Library with proper EMPTY states ("be the first to create one") and a green "✓ working version" badge so the team can see the suite of audios that are actually ready. (4) Share + HARD REQUEST with a note + notify (required requests stay flagged on the recipient's list). (5) Reports on every audio. (6) Relationship / Synergy (combine 2+ charts) and one-click Deep Cosmic Report (report only, any person). (7) New-category creation (some categories like EFT may hold several audios + instructions). (8) Login: log-in-as-anyone kept for Jeff's LOCAL build; the WEB build gives Sol and Jeff their own logins (Colin later). NAMING: full names everywhere (Jeff Parker Love, Sol Ballard, Colin Jacobs, Josh Parini); equal billing, no "founder" slight, never "system/soul" framing. NEXT: Claude Code builds to this brief and deploys to the web so Sol can test.

  Open studio refinements (not yet done): a few structures slightly overran the word cap (founder structure hit 1009 vs the 950 target) so per-structure length tuning is worth a pass; the Studio Spec's later layers are not built yet (comfort vote 1 to 5, A/B variants, multi-person feedback, the leaderboard, push-to-inner-circle, the techniques corpus tradition, Google Sheet tracking, the known-chart accuracy harness). Cloud deploy (Render + R2 + Cloudflare Pages + Cloudflare Access gating on studio.affirmology.ai) is pending Jeff creating the Render/R2 accounts.

  12. Council/loop + relationship engine (added 2026-06-16)

  • COUNCIL + LOOP as PROMPTED ROLES: BUILT as prompts and DEMONSTRATED end-to-end (not yet wired into code). The "feels deeply seen" quality layer. Roles: Sophia (synthesist pre-pass brief across all systems), Chiron (technique plan, see below), Orpheus (mythopoet draft, Audio Openings Framework baked in), Apollo critique loop with four gates (technique-fit, arc-and-cadence, Hypnos subconscious-efficacy, North Star). Prompts: CLAUDE OUTPUTS/Affirmology/Affirmology_CouncilLoop_Prompts_v1.md. Live demo on Sol (v1 generic -> loop -> v2 cleared, showing the lift): CLAUDE OUTPUTS/Affirmology/Affirmology_CouncilLoop_Demo_Sol_v1.md. NEXT CODE STEP: add sophia_brief(), chiron_plan(), apollo_review(), compose() to script_generator.py and verify with a live API render (needs Jeff's go). Acceptance rule honored: script-only, shown, no ElevenLabs.
  • CHIRON, the Technique Healer (added 2026-06-18, prompted role): runs ONCE after Sophia and before Orpheus. Given the person's synthesis + the audio's structure/purpose + the Techniques Library (Affirmology_TechniquesLibrary_v1.md; later the structured "techniques" corpus tradition), it SELECTS and SEQUENCES the fitting techniques and outputs an ordered technique plan (technique | position in the arc | one-line rationale) plus a "left out and why" note. Orpheus weaves the techniques in at Chiron's positions; the build report ("Why this build") prints the plan verbatim. The existing Apollo technique-fit gate and Hypnos gate are KEPT and now CHECK Chiron's plan (fit + adherence), so techniques are designed up front instead of discovered reactively through the loop. Brand rules honored: never names the meditation teacher (heart coherence breath / embodied gratitude induction); sound/frequency layers claimed as felt experience only, beds only where they fit. Before/after demonstration on Jeff's real chart (founder-confidence audio): CLAUDE OUTPUTS/Affirmology/Affirmology_CouncilLoop_Chiron_Demo_Jeff_v1.md (shows the plan and how the script changed). Still prompts-only; wiring into code waits with the rest of the council.
  • PERSEPHONE, the Breakthrough/Transformation guardian (added 2026-06-18, prompted role; name chosen by Jeff). The council's champion of the transformative piece, the counterweight to identify-and-align. Given the synthesis + the audio's purpose, it ensures every audio carries a real shadow-to-gift arc (Gene Keys shadow -> gift -> siddhi), concrete guidance on how to ACT in life, and in deeper programs a genuine breakthrough on what limits the person (drawing on astrology and the rest). Descent-and-return is the archetype: through the dark and back, renewed. It reviews scripts so transformation shows up in ALL of them, not only the align-the-identity ones, and it names Sol's lane in the council (she leads transformational journeys). IS AN ACTUAL APOLLO GATE (Jeff 2026-06-18), not just an advisor: add a "transformation gate" to Apollo's loop so breakthrough is ENFORCED on every script, not optional. DISTINCT FROM CHIRON: Chiron handles neuroscience/meditation TECHNIQUES (the methods, the how); Persephone has the BROADER palette: she ensures the right transformation ARC is applied with NUANCE and CONTEXT for the audio type. The amount of transformation and the WAY it shows up differ by audio type (a morning align-and-go carries a light touch; a deep journey or breakthrough program carries a full descent-and-return). So Persephone is contextual, not a fixed quota. Prompts-only first like the rest of the council; wire into council.py + Apollo's gates with Sophia/Chiron/Orpheus after. Gate prompt drafted in CLAUDE OUTPUTS/Affirmology/Affirmology_Persephone_GatePrompt_v1.md. Origin: Jeff's 2026-06-18 note that the breakthrough piece "almost needs to be a bot itself."
  • RELATIONSHIP ENGINE: BUILT and tested. affirmology-agent/src/affirmology/chart/relationship.py, additive, demo untouched. Shared two-chart synastry engine (HD connection + Penta, Western synastry/composite, Vedic Moon nakshatra + partial Kuta, Gene Keys Venus overlap), reusing human_design.py maps. It is the shared engine under Prometheus + the Eros/Concordia/Hestia/Gaia lenses, per the council doc. Spec: Affirmology_RelationshipAgent_Spec_v1.md. TODOs: full Vedic Ashtakoot, Western interpretation/weighting layer, wire into Prometheus and lenses (Eros-for-clients first).

  13. Studio LIVE + live-use punch list (2026-06-16)

  Studio is LIVE at studio.affirmology.ai (gated for Jeff + Sol; the affirmology-studio-api Worker routes to the laptop funnel; SSD persistence). Jeff rendered his own audio: the story "sounds like me" (the North Star showing up), voice clear, good music balance. The single prioritized build queue is Affirmology_StudioChangeList_v1.md. Key items: - P0 CRITICAL chart-integrity bug: a report crossed two charts (placeholder "render"/2000-01-01 in the header/blueprint/section-labels; Jeff's real data in the prose). Cause: people library not wired (so the render used the placeholder identity) and report vs script read different chart sources. Fix: wire the four real charts so a selected person passes real birth data; report + script from ONE chart computation; never fall back to a placeholder (fail loudly); purge the "render" job. Oracle: Jeff = Virgo Sun, Cancer Moon, Capricorn Rising, MG, Sacral, 2/5, Single, Gate 47, Life Path 22. - P0 iterative composer: the script step is one-way and loses the user's notes. Keep notes editable after generation, regenerate as versions, render as a separate explicit step. The conversational composer from the mockup. - Live audio feedback: music bed must cover the FULL audio length (loop/extend/fade tail); cadence, let key recognitions breathe instead of being rattled past (the council loop's job). - Voice-note feedback; feedback reachable from EVERY screen and accepting a long pasted transcript; co-created length (stop defaulting to the demo's ~6 min); named agents speak in the UI (Sophia/Orpheus/Apollo), shown in the v3 mockup. - Craft: structure-aware endings, and the openings AND endings are crafted by Orpheus as bookends (never a placement checklist). Docs: Affirmology_AudioOpenings_Framework_v1.md, Affirmology_BeforeBed_AudioStructure_v1.md. - Design source of truth for the UI: CLAUDE OUTPUTS/Affirmology/Affirmology_StudioUI_v3.html. Agent design: Affirmology_AgenticCouncil_Design_v1.md, Affirmology_InsightAgents_Design_v1.md. Build brief: Affirmology_StudioBuildBrief_v1.md. Change queue: Affirmology_StudioChangeList_v1.md. - Ops: Affirmology_StudioDeploy_Runbook_v1.md (Phase 1 done; Phase 2 = Render + R2 later). Living user guide with changelog: CLAUDE OUTPUTS/Affirmology/Affirmology_StudioGuide_v1.html (link in the Studio nav). Onboarding any chat to the system: Affirmology_SystemInterface_Guide_v1.md. - Before sending to Sol: land the two P0 fixes (chart-integrity + iterative composer), refresh, then invite her. Security: disable the workers.dev URL on affirmology-studio-api so the gated domain is the only way in. - BUILD ORDER now (per Jeff 2026-06-16): (1) council+loop prompted roles [done as prompts+demo; code wiring next], (2) Blueprint Insight Agent [feeds Sophia, opens Studio composer], (3) wire relationship.py into Prometheus+lenses, (4) Transit & Timing, (5) Potentials & Recommendation, (6) Echo + trend agents. - Kickoff-week client work (2026-06-16): 6 scripts in Affirmology_KickoffWeekSuite_v1.md, deep blueprint report Affirmology_CosmicBlueprint_Report_v2.pdf, analysis, Penta hiring map. All script-only, no renders.

  Build-order step 2 (2026-06-16): Blueprint Insight Agent

  BUILT as a prompted role and demonstrated on Jeff (ranked "what matters most" + convergences/tensions/gifts + suggested audio angles). Spec + demo: CLAUDE OUTPUTS/Affirmology/Affirmology_BlueprintInsightAgent_Spec_v1.md. Output is a BRIEF, not an audio. Feeds Sophia and opens the Studio composer. Code wiring (ahead of Sophia in compose(), and into the composer screen) pending the live-render step. NEXT: step 3, wire relationship.py into Prometheus + the lenses (Eros-for-clients first).

  Build-order step 1 WIRED + LIVE-TESTED (2026-06-16)

  The council/loop is now CODE, not just prompts, and embedded at the creation entry points: - affirmology-agent/src/affirmology/agents/council.py: sophia_brief(), apollo_review(), compose() (Sophia -> Orpheus -> Apollo loop, writes council_transcript.md). Additive. - script_generator.generate_script() gained an additive council_context="" param (empty by default, so the LOCKED DEMO path is byte-for-byte unchanged). - cli.py gained --council; both script-gen call sites route through compose() when set. - Studio pipeline.run_script() now passes --council, so every Studio creation runs the loop. The demo backend never passes it, so the demo is untouched. - LIVE TEST (real Anthropic API, model claude-sonnet-4-6) on Sol: Sophia + Orpheus + Apollo ran; Apollo passed all 4 gates first draft; output in CLAUDE OUTPUTS/Affirmology/Affirmology_CouncilLiveTest_Sol_v1.md. ElevenLabs voiced a 62-word opening clip OK (Sol_CouncilVoiceTest_opening.mp3). No full render. - Sandbox note: the engine venv is a Mac build (broken on Linux); tests here used system python3 with pip-installed deps. Full --audio (ffmpeg mix + QC + WeasyPrint PDF) should run on the Mac.

  13. Studio change-list progress (working Affirmology_StudioChangeList_v1.md top-down)

  • P0.1 chart-integrity bug: FIXED in code (2026-06-16). Root cause confirmed: affirmology-studio/api/pipeline.run_audio shelled the CLI with a PLACEHOLDER identity (--name render --date 2000-01-01 --location n/a), so the Sacred Audio Report recomputed a placeholder chart (Cap Sun, Scorpio Moon, Aries Rising, GK38, LP4, 1/4 Split) while the script was the real person. Fix: run_audio(job_id, birth, ...) now requires and uses the job's OWN stored real birth data; jobs._run_audio builds BirthInput from the stored row and fails loudly if birth data is missing (no placeholder fallback). The people library already seeds the four real charts and the script path already used them, so report + script now compute from the same real birth data (deterministically identical chart). Verified: Jeff's chart from one source = oracle (Virgo Sun, Cancer Moon, Cap Rising, MG/Sacral/2-5, LP22, Gate 47). Still to do on the Mac: (a) run one Jeff --audio render and eyeball that the PDF header/blueprint match the script (WeasyPrint runs there, not in this sandbox); (b) purge the leftover "render" job row from the live jobs DB (harmless now; new ones can't be created and would fail loud).
  • P0.2 iterative composer: NEXT.
  • VEDIC GAP (flagged by Jeff 2026-06-16): there is NO Vedic/Jyotish chart engine in src/affirmology/chart/. The Western engine is tropical only. The only Vedic code is the partial Moon-nakshatra + 3-of-8 Kuta inside relationship.py (synastry). Parashara is designed in the council doc but unbuilt. Any "Vedic" in the reports so far was hand-computed, not engine-driven. Proposed: build a real chart/vedic.py (Lahiri sidereal positions, Moon nakshatra/pada, Vimshottari dasha, D9 Navamsa, key yogas) so Parashara has real data. Queued; sequence per Jeff.

  Vedic incorporated into the SYNTHESIS layer (2026-06-16, per Jeff)

  Jeff's call: not a separate Vedic silo yet (custom Vedic audio paths come later), but Vedic should speak when SYNERGIES appear across systems. Done: new src/affirmology/chart/vedic.py computes the Lahiri sidereal layer from the chart's already-computed tropical longitudes (no re-geocode): sidereal rashis for Sun/Moon/Ascendant, Moon nakshatra + pada + Vimshottari lord, gana, nadi. vedic_facts_text() is now fed into council.sophia_brief() so Sophia can let the sidereal sky confirm or complicate a convergence. Verified facts: Jeff sidereal Sun Leo, Moon Gemini/Ardra (lord Rahu), Asc Sagittarius; Sol sidereal Sun Capricorn, Moon Gemini/Ardra (lord Rahu), Asc Pisces. Notable synergy: BOTH founders share a sidereal Moon in Gemini, nakshatra Ardra, Rahu-ruled, reinforcing their shared tropical Cancer Moon. Live full-Sophia weave not shown here only because one Sophia call exceeds this sandbox's 45s shell cap; run on the Mac to see it woven. Full Vedic (dashas, Navamsa, yogas) + Parashara + Vedic audio paths remain queued.

  Vedic = part-now / part-later project (2026-06-16, Jeff)

  Plan: CLAUDE OUTPUTS/Affirmology/Affirmology_VedicEngine_PhasedPlan_v1.md. PART NOW (done): synergy-level sidereal facts (chart/vedic.py) feeding Sophia. PART LATER (phased, heavy): L1 Vimshottari dasha, L2 divisional charts + yogas, L3 strengthen the WEAK Vedic corpus (required before Vedic-led generation), L4 Parashara + dedicated Vedic audio paths. Every phase must pass a known-Vedic-chart fixture before shipping (same discipline as HD vs Josh). No Vedic-led audio yet. Keep Jeff updated here + in the change list.

  P0.2 iterative composer DONE (2026-06-16)

  Studio composer is now iterative (minimum viable, matches the mockup intent). Changes: - web/index.html: added an editable "Notes for this version" textarea (composeNotes) + a "Regenerate as new version" button (reviseBtn) in the composer rail. The script keeps showing in the chat alongside the notes; Render stays a separate explicit step. - web/app.js: refactored the compose call into a shared runCompose() used by both Send (append a change) and reviseFromNotes() (edited notes replace the brief). Every compose creates a NEW version (backend _next_version increments per person+category, confirmed). Added briefKey/persistComposer/syncNotes; the working brief is saved to localStorage and restored in bootChat, so navigation/reload loses nothing. setComposeBusy disables both buttons during a draft. - Verified: node --check web/app.js passes; wiring consistent; backend versioning confirmed. Full browser click-through is a Mac check (live FastAPI + browser not runnable in this sandbox). - P0 (chart-integrity + iterative composer) now COMPLETE. Next: Priority 1.5 (named agents in UI, feedback-everywhere + voice notes + pasted transcript, co-created length) and the live-audio fixes (music covers full length, structure-aware/crafted endings, cadence) which lean on the council loop already built.

  14. Tracked PROJECTS (long-term, flagged 2026-06-16)

  • REPORTING SYSTEM OVERHAUL (all reports, not just the cosmic one): typography/readability (embed real brand fonts, larger sizes, full-cream contrast), kill deadspace (on-brand Hellenistic + constellation SVG visuals, big chart-drawn mantra pull-quotes, a few more agent voices), apply across every report we generate. Plan: CLAUDE OUTPUTS/Affirmology/Affirmology_Report_Redesign_Plan_v1.md. Interim readability already applied in the Colin report (Liberation fonts embedded, sizes up, agent voices enlarged) and ready to backport.
  • COUNCIL CHARACTERS / AVATARS: a marble-and-gold Hellenistic visual character for each named agent (Sophia, Hermes, Parashara, Pythagoras, Prometheus, Athena, Orpheus, Apollo, Hypnos, Echo, Eros, Concordia, Hestia, Gaia), used beside their voice in reports and in the Studio composer. Brief: CLAUDE OUTPUTS/Affirmology/Affirmology_CouncilCharacters_Project_v1.md.
  • COUNCIL CHARACTERS - VOICE + HUMOR SUITE (2026-06-18, Jeff asked Cowork to "make a whole suite of characters"): rich personality, humor profile (or pointed non-humor), speech signature, inter-agent banter, user-facing manner, and sample lines for all 16 oracles. Doc: CLAUDE OUTPUTS/Affirmology/Affirmology_CouncilCharacters_VoiceProfiles_v1.md. KEY GUARDRAIL baked in: personalities show in the composer/council transcript/Hermes chat, NEVER inside the Sacred Audio (the audio stays pure first-person). Humor is premium, room-reading, dials to zero on tender topics. Nice touch: Pythagoras in-character REFUSES name-numbers without a confirmed full name (mirrors the numerology accuracy rule). HUMOR CALIBRATION (Jeff 2026-06-18, important): they must NOT be funny all the time and must not sound like one person with different names. Added a "humor dial" to the doc: most of the council is LOW or near-zero humor (Sophia/Hypnos/Parashara/Gaia near zero; Apollo/Persephone/Athena/Concordia low-dry; Hermes/Eros/Echo/Prometheus high-but-situational). Personality comes primarily from VALUES + WORLDVIEW (what each notices first on the same chart), not jokes. Differentiation test baked in: swap two agents' names and if you can't tell, fix by changing what they VALUE, not the joke. Default to the low end; heavier/tender moments lower every dial. Hermes's live prompt updated to "humor is occasional, not default" (helpful/warm first). Visuals (marble-gold avatars) coming per character. WIRED NOW: Hermes's voice+humor folded into the live /api/hermes system prompt (api/main.py), so Atlas chat has his personality on first test. TO WIRE NEXT: per-role VOICE blocks into agents/council.py so Sophia/Orpheus/Apollo/Chiron/Persephone speak in character in the transcript + "Why this build" report (after Code's corpus run, since council.py is in the shared engine repo). ROSTER LOCKED (2026-06-18, Jeff): the five tradition experts are Western astrology = SOPHIA (she also synthesizes; the lead, most client-facing oracle), Gene Keys = ATHENA, Human Design = PROMETHEUS, Vedic = PARASHARA, Numerology = PYTHAGORAS. HERMES is the assistant + ROUTER (front desk who hands questions to the right expert), NOT a tradition specialist. (My first VoiceProfiles draft wrongly made Athena a "strategist" and Prometheus the "relationship engine"; the older Affirmology_CouncilCharacters_Project_v1.md brief had Athena=Gene Keys + Prometheus=Human Design RIGHT all along; the only real change is Western astrology moved Hermes -> Sophia because Hermes became the Atlas assistant. Both character docs corrected.) HERMES CLIENT-CONCIERGE = PHASE 2: today internal-team only; later the client's concierge/router, gated by a new CLIENT permission tier (own chart + audio only). Captured in Affirmology_Atlas_Overview_v1.md section 3.
  • GAMIFICATION & RITUAL DESIGN - LIVING DOC STARTED (2026-06-20, Cowork): CLAUDE OUTPUTS/Affirmology/Affirmology_Gamification_v1.html. North star = mechanics must serve the transformation (Subconscious OS frame: installing updates / being initiated, not earning points); earned-not-bolted-on, ritual over reward, no shame mechanics, cold-to-warm, and ALWAYS corpus-true (gamification shapes pacing/presentation/progression, never the truth of a reading). FLAGSHIP PARKED CONCEPT = CHOOSE YOUR SPIRIT GUIDE: after a user listens to a meaningful amount of base audios they UNLOCK the ability to pick a Spirit Guide who becomes their customer-facing personality/relationship/presentation layer (their own Hermes/Sophia). KEY REFINEMENT (Jeff, locked thinking): nothing changes architecturally - HERMES STILL LINKS ALL of it (routes to the oracle specialists, corpus grounds the answer); the guide is purely the human-facing voice Hermes brings forward for the message + relationship. Guardrail = guide picks wording/warmth, the oracles + corpus still supply the truth (same rule the chat follows). The unlock gate does 3 jobs at once: drives base-audio consumption, makes the guide feel earned, prevents day-one choice overload. PHASING: v1 = use the EXISTING character suite (Sophia, Atlas, council) as guide options, ships inside what exists; v2 = a large multi-tradition archetype library (male + female) as the rich expansion. This maps onto HERMES CLIENT-CONCIERGE = PHASE 2 above (the client's gated concierge/router). The doc also seeds a progression spine (Arrival -> Initiation -> The Guide -> The Councils -> Integration -> Mastery), seed mechanics (constellation-fills-in instead of streaks, initiation track, integration quests = tapping/gratitude/chatbot, sky-timed rituals, council unlock ladder, create-your-own journeys, the Affirmologist community path, milestone ceremonies), and themed TRACKS (Initiation, Wealth/Abundance, Love, Purpose, Rest, Shadow/Breakthrough). ONGOING: keep adding gamified rituals + tracks to this doc. CONSTELLATION SYSTEM (Jeff 2026-06-20, added to the doc): the sky-constellation + lights-unlocking theme is the spine of the whole unlock system; a user holds SEVERAL constellations at once (some filling in PARALLEL), each tied to a theme/tradition; COMPLETING a constellation unlocks something for it, headline reward = ACTIVATING a Wisdom Council member as a guide (the constellation you fill brings the guide to life), plus "that and more" (a deeper reading, a track, a ceremony). So the Spirit Guide unlock runs ON the constellation mechanic, not a separate progress bar.
  • KNOWLEDGE/DELIVERY LAYER (2026-06-20, Cowork): TWO separate jobs disentangled. (1) JEFF'S find-it-fast layer = OBSIDIAN (his installed PKM; Notes is weaker). New curated map-of-content over the reports: CLAUDE OUTPUTS/Affirmology/Affirmology_Reports_Obsidian_Index_v1.md (point a vault at the Affirmology outputs folder; tags/search/graph; links to the HTML deliverables; PROJECT_STATE stays the narrative source of truth, the index is the browse layer). Since most deliverables are brand HTML and Obsidian's native unit is Markdown, the pattern is an MD index + short MD companions linking to the HTML. (2) SOL/COLIN delivery + idea-capture = an ATLAS feature, NOT Obsidian (single-player). Brief: CLAUDE OUTPUTS/Affirmology/Affirmology_AtlasReportsHub_Brief_v1.md. Direction (Jeff): a REPORTS spot on the Hermes side where Sol/Colin browse pushed reports and easily give ideas (TEXT now, VOICENOTES when enabled), plus an email nudge in parallel. Rides on rails ALREADY built (C16 deliveries/inbox + send_to_person + tiers + cross-device Hermes history + voicenotes v1). Phasing: v1 text + Reports tab + Open-full link + tagged responses + Jeff Responses view (additive, demo untouched); v2 voicenotes via Whisper/on-device Parakeet, GATED ON APPLE DEVELOPER TOOLS ("soon"); v3 Hermes synthesizes responses per report. Hosting the styled HTML link = the R2 public-media (C10) dependency. EMAIL NOT SENT yet (Jeff holding; many such docs to come, build the mechanism first).
  • FILE ORG CLEANUP - PLAN DRAFTED, NOTHING MOVED (2026-06-20, Cowork): Jeff flagged the nested CLAUDE OUTPUTS/Affirmology folder as redundant/confusing (root cause = a line in his PERSONAL GLOBAL instructions that writes outputs to CLAUDE OUTPUTS/<project>/, which fires even though /CLAUDE/AFFIRMOLOGY is already the project). Plan: Affirmology_FileOrg_Plan_v1.md (root). DO NOT reorganize or move files without Jeff's sign-off on the target structure. Key facts from the dependency scan: only 15 files reference the CLAUDE OUTPUTS path (10 root prose incl. PROJECT_STATE, 3 inside the folder, and 2 RISKY CODE files that hardcode it: affirmology-studio/api/hermes_tools.py + affirmology-agent/src/affirmology/corpus/craft_seed.py); Obsidian vault is rooted at the PROJECT ROOT (in-project moves won't break it); root is NOT a git repo (use a zip backup + symlink pointers). Safe method = move + leave a SYMLINK + _MOVED.md stub at the old path so old paths still resolve, update the 2 code paths explicitly, find/replace prose refs. Recurrence fix = add a "no CLAUDE OUTPUTS folder here" convention to the project CLAUDE.md AND change the global instructions line (Jeff does the global edit). Phasing: Phase 0 new files go to the new structure now (zero risk); Phase 1 retire the CLAUDE OUTPUTS folder; Phase 2 sort the ~252 loose root docs.
  • INVESTOR VIDEO - NEW SCENES (2026-06-21, Cowork overnight): added the StoryBrand "why this matters" GUIDE scene + rebuilt Joe/Monika beats. Gallery: CLAUDE OUTPUTS/Affirmology/Affirmology_VideoScenes_OvernightGallery_v1.html; clips in Affirmology_VideoEditKit/{rooftop_guide,broll_why,joe_alignment,monika_arc}. LOCKED DECISIONS (Jeff): (1) GUIDE = the existing blond male friend (he carries the scene; Joe barely talks, his beat is "why does astrology even matter?"). (2) SEQUENCE keeps ALL existing scenes; INSERT the guide convo. Two-phase guide: Phase-1 guide introduces the WHY (new scene), Phase-2 guide answers the HOW (existing friend-hands-phone + couch). Order: NEW friend conversation (why) -> Joe at home w/ books+report frustrated (linear brain) -> friend hands phone -> friend on couch as Joe gets on Affirmology -> love affair. (3) Guide scene LOCATION = ROOFTOP at golden hour (people in bg), NOT gym/cafe (Joe is already established as a workout beast via ice bath/run/beach, and the Aurum cafe is reused for the Monika reunion). Generated: rooftop two-shot (friend talks/Joe listens) + reverse (Joe talks) + OTS singles + tight friend CU + alt takes (real shot/reverse coverage; singles must be DIFFERENT ANGLES/OTS or the model duplicates/erases a person). (4) Cutaways under the friend's VO: mountain-temple seeker, cosmic astrological CHART-IN-THE-SKY, and a SOUL-BIRTH body-of-light. JEFF LOVES the cosmic chart + soul-of-light look = adopt as a DEMO-VIDEO visual lane. (5) Joe BEACH-RUN replaced by a MIAMI DOG-WALK: coastal park, over-ear headphones, gold constellations forming on his chest (alignment motif), bay + Miami Beach skyline across the water. Also: swap one of the two JOG scenes for a dog-walk (pending). (6) MONIKA earned arc (suite + repetition, not one trick): keep her app "blocked area" chat + business-success + bath; ADD wealth-codes AHA (engagement) + a varied-doors montage (morning activation, athletic, heart-coherence car moment, EFT as one quick flash, before-bed) with on-screen labels; headphones throughout; app screens EMERALD+GOLD Affirmology branding (NEVER Aurum; Aurum = the physical studio only). On-screen guide words (overlays, system already built): PURPOSE / WHY YOU KEEP HITTING THE SAME WALL / HOW YOU'RE WIRED / HOW TO BREAK THROUGH. Drop the "microdose" door (investor optics). STRATEGY note that drove this (from another chat's manifesto doc): lead with cosmic-blueprint recognition BUT make the self-realization purpose explicit >=2x so a cold viewer/investor never reduces it to "an astrology app" (answers Sol's concern). EM-DASH-FREE. CREDITS: Higgsfield ran DRY at the end of this batch (balance ~1.18; "out of stills"). All 20 overnight clips finished before it drained. QUEUED for when credits return: (a) redo the temple cutaway with the FRIEND as the climbing seeker (not a generic monk); (b) extend the cosmic-chart/soul-birth look for the demo; (c) the COUNCIL OLYMPUS-ORB scene Jeff requested (core council, not wisdom circle, gathered around a big rotating light-filled orb looking into someone's life; a premium contrast to a mechanical agent/oracle diagram) - HELD pending credits per Jeff. Jeff likely needs the bigger Higgsfield account. (d) REDO the Monika heart-coherence clip: current one shows her in a MOVING car with headphones on and hands in her lap (reads as distracted driving = unusable); regenerate as a clearly PARKED/stationary car or move it to a still setting (window seat at home, couch). All other Monika doors + dog-walk + rooftop coverage are usable. CREDIT/PLAN NOTE: Higgsfield Plus shows "365-day"/"7-day" UNLIMITED perks for some models (Nano Banana, Kling 3.0, etc.) BUT they're tied to a promo purchase window ("buy until ~June 20") and marked "available on web"; the MCP/Cowork generation path bills the METERED credit pool (image=1-2 cr, Kling video=7.5 cr each), which is why we hit a wall at ~1.18 cr despite the "unlimited" wording. Jeff to verify exact entitlements on higgsfield.ai directly; options = top-up (500/$26 … 4000/$190, 90-day expiry) or upgrade Ultra ($129/mo, 3000 cr) + auto-refill.
  • COUNCIL CHARACTERS - VISUAL BUILD (2026-06-18, Cowork, big session). Generated marble-and-gold Hellenistic busts for the whole council via Higgsfield Nano Banana Pro (model runs as nano_banana_2). LOOK DNA (locked): living marble-and-gold statues with faint inner constellations and fine gold kintsugi veins; TWO-STATE EYES, glowing gold at rest and warm amber when speaking, NEVER blank carved eyes; clothed and posable; STONE-TONE DIVERSITY (cream, honey, bronze, black-and-gold Nero, verde green) so the pantheon is not all-white; deep emerald + gold + cream palette; square chest-up busts, distinct silhouette per character. LOCKED so far: ATLAS = the whole SYSTEM personified (not just the app; Hermes is a smaller in-app messenger). Astrologer-sage, cream-white marble (stays white as the Western-facing anchor), glowing gold eyes, a bold golden armillary compass-globe on a sweeping orbital swoosh; warm SMILE and LAUGH added to his range. UPDATE 2026-06-18: BALD is now his PREFERRED primary look (full beard kept), with the HOOD kept as a variation; a 4K bald character model sheet was built. KINTSUGI DECISION (Jeff, final): LOCKED PRIMARY Atlas = bald with a MEDIUM (small, tasteful) gold kintsugi patch on the skull and CLEAN robes (job 949ba972); enough character for interest, subtle enough for investors. The bunched-up hood collar is OUT (it read like a bolted-on attachment); his default is the clean drape, and when he hoods up occasionally the hood must be rendered INTEGRATED, not a separate piece. The bolder gold-crown is parked as a future "evolution"/premium variation. Suite: CLAUDE OUTPUTS/Affirmology/Affirmology_Atlas_BaldSuite_v1.html.
  • WISDOM CIRCLE - first wave generated (2026-06-18, all liked by Jeff): RUMI (Persian Sufi, honey-amber, dervish cap, reed flute/rose), SIDDHARTHA (Buddhist, honey-cream, ushnisha, lotus), LAKSHMI (Hindu, bronze-gold, lotus crown, coins of light), ISIS (Egyptian, cream-bronze, solar-disc crown, wings, ankh), IX CHEL (Mayan, jade-bronze, crescent-moon, weaving). ALSO LOCKED (2026-06-18): EARTHKEEPER (North American Indigenous man, the Eagle-and-Condor unity prophecy, lighter-hair take, job 09bd50b7; respectful/general, no specific sacred regalia), TALIESIN (Celtic druid-bard-seer, verde-green marble, torc/oak/knotwork, job 024e36ba), and "YOUNG" the Jungian depth-psychologist rendered in MODERN dress (Swiss-professor look, suit + spectacles + mandala, job 7ac7df3a). KEY VISUAL RULE (Jeff): the MODERN-era thinkers (Young, neuroscientist, Campbell, Wilber) get MODERN clothing and read contemporary, marble statues in suits/glasses; the ancient/mythic figures stay robed. NEUROSCIENTIST: LOCKED as the modern Black man (job b6ebe120), modern dress, brain drawn as a constellation star-map (Jeff loved it; the South Asian woman 612f0edf is the alt). The brain-as-constellation motif ("the sky within") is a strong on-brand idea for the subconscious-OS frame. LOGO CANDIDATE (Jeff likes it, 2026-06-19): the mark that emerged on the in-app mockups = a GOLD CIRCLE CONSTELLATION with stellar swooshes on deep emerald. No official logo existed; Jeff approves this direction as the working logo. Appears on app screens: joephone 275c370b, friend-hand 8fab8268, app-speaking edb74fa6, showapp 8ca8fb64. TODO: isolate/render a clean standalone logo lockup from it. ALSO LOCKED: the coffee-shop POV of handing Monika the Affirmology phone (showapp, job 8ca8fb64/anim d950a585) = the "showing her the app" beat. OPENING SEQUENCE - ROUND 2 additions (2026-06-19): NEW scenes generated + many ANIMATED (kling): yoga-class opener (anim 01871e90), friendly walk-up (anim 735e2e34, re-done less flirty), friend-shows-app (still 0edef6c9, anim 3abfc857), GROUP ecstatic dance (still 07ab23b0 re-done not-romantic; earlier couple-dance 14675b8c/anim 6b8623cc rejected as too romantic), fun-life: headphones-yoga f668f406, breathwork-circle e88e03b4, cold-plunge 33675ba3, joe-park 94abfc20, fun walk. THE OVERLOAD DESCENT (Jeff loved it): SC1 gibberish now drops INTO Joe's mind - dizzy glyph spin (anim 23e51b90), tangled cosmic mind (anim ee98e51a), vortex tunnel brain->cosmos (anim 81af5725). Concept: surfaces his LIMITING BELIEFS ("I'm not enough / I'm behind / everyone gets it but me / this isn't logical") as Remotion words, which his Sacred Audio then answers line-for-line = setup/payoff = the whole thesis in 10s. Doc: Affirmology_OverloadConcept_v1.html. DIRECTION: keep Monika's gibberish near FULL length (the descent carries it, don't over-trim). RULE reinforced: Joe/Monika stay AMBIGUOUS/platonic - no flirty, no romantic, no touching. Gratitude practice belongs in MONIKA's app-use. SC5 = KEEP overlapping his/her voices + Remotion caption "Same music, customized audio script". Phone app+promise scene added (Joe+phone 275c370b, insert 648919cf). NEXT: full ANIMATED assembly v5 (swap animated clips into all beats, new opening, overload descent, disclaimer fun-life montage, captions). OPENING SEQUENCE CUT v3 = current (2026-06-19): CLAUDE OUTPUTS/Affirmology/Affirmology_OpeningSequence_v3.mp4 (2:20). Built AUDIO-FIRST per Jeff: full dialogue voiced (Inworld, Monika=Olivia, Joe=Ethan, Friend=Liam, Oracle=Serena, Narrator=Evelyn), silence-trimmed, stitched to a master VO with pauses for non-verbal beats + overlapping his/her stars audio + music bed, then sped atempo 1.30 to hit 2:20. Master VO: Affirmology_OpeningSequence_MasterVO_v3.m4a. Video = Ken-Burns animatic of painterly stills cut to the master track (segplan in outputs/assembly). Timed shot list: Affirmology_OpeningSequence_TimedShotList_v1.html. FIXES applied: walk-and-talk opener added (was missing), Monika 3-arm artifact fixed (new walk 572bf53c), montage line now chart-specific "Gate One, create not compare" (8659abbb), app+promise scene added (Joe+phone 275c370b, phone insert 648919cf), chatbot branded Affirmology not "Aurum Oracle" (34e51e73), jog single-Joe fixed (6648672d), gibberish rides into babble overload. NEXT PASS: swap hero beats to the kling-animated clips, layer Remotion on-screen text, optional ElevenLabs final voices. Lesson: build the master VO first and cut picture to its timestamps; do not force fixed-length clips. VIDEO CONTINUITY RULE (Jeff, 2026-06-19): whenever an avatar is LISTENING to Affirmology, HEADPHONES must be visibly on, every time. PREFERENCE: over-ear headphones (easier to see on screen); earbuds acceptable. GREENLIT (2026-06-19): Jeff loves the storyboard v2 frames + scratch VO; full motion-and-music pass approved. Monika's abundance payoff = HER BUSINESS BLOWING UP (job 3ed14d37). MORE RULES (Jeff): (a) Scene 2 friend is IN PERSON, not a text (rhymes with Joe handing it to Monika in person in SC4 = app spreads person-to-person). (b) The alignment montage is a LOVE-AFFAIR montage but the love affair is with HIMSELF (self-recognition); it makes him magnetic. (c) The Joe/Monika connection is AMBIGUOUS: they NEVER hold hands or kiss; platonic-or-romantic is the viewer's call. (d) Added a tasteful candlelit BATH listening beat for Monika (over-ear headphones, no nudity, shoulders up; job aa6127f2) as her love-affair mirror to Joe's montage. Scratch audio rendered: music bed 0537817b (Sonilo 45s), Joe Sacred Audio clips (Ethan) 796e86a0/23916773/7152d70c/36f0b46d, narrator (Evelyn) ed82a1e7, Monika clip (Olivia) a8e81a33, Atlas e05e8626, disclaimer f15470b6. SC5 dual overlapping clips: his (Ethan) c15cba56/1034d57c, hers (Olivia) 83ea5744/be0794bb. FULL MOTION PASS DONE (2026-06-19, kling3_0_turbo, 5s/720p, ~7.5 cr each): babble a41381cd, couch 18d68a6c, eureka 44add432, jog a65d4816, kitchen ab4aadc3, juice 5574b142, stars 8eb43be9, bath ae5e671f, EFT 4238d348, business 81924a68. Galleries: Affirmology_MotionPreview_v2.html (all 10), Affirmology_AvatarSequence_Storyboard_v2.html (stills+audio). Remotion overlays coded: CLAUDE OUTPUTS/Affirmology/Affirmology_RemotionText/. SCRIPT MODS this session: SC1 Monika babble = denser Gene Keys gibberish; SC5 runs LONGER w/ several parallel his/her text pairs + overlapping male+female audio clashing over one music bed. NEXT: assemble clips to the locked audio + Remotion text in an NLE/Remotion; extend SC5; then build the Atlas-voice thesis trailer + rest of film. VIDEO PRODUCTION PIPELINE (Jeff direction, 2026-06-19): AUDIO-FIRST. Lock VO + music timing first so it dictates exact video clip lengths, then generate footage to fit (with handles/variations per scene to extend in the edit). Do the WHOLE sequence, not parts, with a few variations per scene. ElevenLabs is NOT available as a connector in Cowork; voicing done via Higgsfield Inworld TTS (70+ voices; Atlas = "Hades", disclaimer = "Brian") + Sonilo music; ElevenLabs stays an optional final swap on the Mac engine. Atlas trailer VO rendered (job e05e8626), disclaimer VO (f15470b6). SC5 dual-audio scene must run long enough to clearly show TWO different audios + two sets of on-screen text playing at once. Storyboard frames (painterly, anchored to ref 2a7421c8): babble f3f6bd61, couch-books 67e29bef, eureka-earbuds 37a3471a, jog c1f99f4c, kitchen(+earbuds) c3f6d43a, juice-shop eb4de27d, stars 8a445f43, EFT 75504b15. INVESTOR VIDEO - HUMAN LAYER (2026-06-19): the customer-world look is LOCKED as PAINTERLY SEMI-REAL illustration (warm, between photo and illustration, soft brushwork, emerald+gold light) so the human world harmonizes with the painted marble-oracle world. Two avatars are invented original ACTORS (not real likenesses): MONIKA (Convinced Woman - radiant warm woman early 30s, long wavy brown hair, sage-green athleisure) and JOE (the Seeker - fit friendly man mid 30s, short beard, charcoal athleisure). Locked style/character anchor = the yoga-meeting two-shot outside "Aurum Wellness Collective" (aurum = gold, nice brand tie). Style-test sheet: CLAUDE OUTPUTS/Affirmology/Affirmology_MonikaJoe_StyleTest_v1.html (option 2 = painterly chosen). STORYBOARD TO BUILD (Jeff's rough arc, applied to Colin's submitted idea): (1) Monika+Joe meet after yoga class. (2) Joe gets the app; scenes of him walking his dog in the park with headphones, listening before bed, learning astrology. (3) They talk again, Joe is fluent in the lingo, Monika is impressed, he gets her on the app. (4) A ~45-min Gene Keys journey: same music/time, different voice+content, outside lying under the stars. (5) Monika happily using it: EFT tapping at home, working through issues with the chatbot, abundance theme, receives a financial blessing. Build each beat in the locked painterly style. NOTE: Oshun final image (cowrie-shell crown + gold bokeh sky) is the one Jeff wants but it only arrived as inline previews, not a hostable file; needs a FILE attachment to host + lock into sheets (until then sheets fall back to e5a1a470). BANDIT (Jeff's dog, 2026-06-19) - HANDLE WITH CARE: the real dog was named GAMBIT, and he passed away on Jeff's 40th birthday. "Bandit" is the in-app character name; Gambit is the soul behind it. This is a memorial/tribute woven into the app as a rare recurring easter egg. Treat with warmth and respect always. He is a character: a senior Boston Terrier rendered as a marble-and-gold "God of the Animal Kingdom," but his real nature is TRICKSTER / comic relief. The bit: he's fully convinced he's a cosmic deity while being the council's clown; ~15 personalities, one of which takes itself dead seriously (the god persona). Primary image = job 25c16a27 (take 1, Jeff's pick); alt = da39d918. Built from his photos, reusable Higgsfield media refs: 75fd789e (front, beads) + 76899353 (chair). Jeff may send more photos to tweak. Sheet: CLAUDE OUTPUTS/Affirmology/Affirmology_DogGod_v1.html. RENDER QUEUE (Higgsfield reconnected 2026-06-19 ~02:00; dog-god done): (1) DOG-GOD - a "deity of the animal kingdom," Boston Terrier head (Jeff's senior dog, already wears rudraksha mala beads, serene wise half-lidded eyes), as a living marble-and-gold Anubis-class god statue, beads carved in marble+gold, divine crown/halo, glowing gold eyes, emerald starfield, epic not cute. Reference photos prepped: outputs/dog_IMG_7804.jpg (front, beads) + dog_IMG_8986.jpg (in green egg chair). (2) OSHUN SPARKLE - take the EXACT locked Oshun (job e5a1a470, untouched figure + water) and add the drifting gold bokeh/sparkle ONLY into the dark empty background void (the look from his cowrie-crown bokeh reference). Prior regen attempts (516a73cb etc.) subtly changed the figure, which Jeff rejected; preserve the figure exactly (compositing is an option if regen keeps altering her). Higgsfield MCP (server 75975218) disconnected mid-session; reconnect then fire both. ATLAS CARD RULE (Jeff, 2026-06-18): Atlas's default/primary look uses the NON-HOODED clean robes, NOT the bunched-up cowl. The hood is a secondary variation only. Use the non-hooded clothes on his character card and anywhere he's the lead. VISUAL BIBLE built (2026-06-18): CLAUDE OUTPUTS/Affirmology/Affirmology_VisualBible_v1.html consolidates EVERYTHING in one file with a jump menu, Heroes (2 + 4K model sheets), Core Engine (16), Hermes morphs (5, rarity-tagged), Wisdom Circle (13 incl. both Oshun takes), Founder (Sol), Worlds (4). NEXT (possibly overnight): build individual CHARACTER CARDS to the BuildStandard, then train Souls for locked heroes. Only Wisdom seats not fully locked: the Sound Healer (Polynesian) and which Oshun variant. SOL BALLARD ("Mindset Ninja", founder) = LOCKED (2026-06-18) as a LIVING MARBLE-AND-GOLD STATUE built from her own photos (face media 5b03f465 + hair media 4a0ed5e1): warm bronze-cream marble, her exact joyful smile + full cheeks, big dark natural curls, a MODERN stylish dress CARVED in marble with gold-thread trim, clean marble face with NO facial kintsugi, gold kintsugi veins only on the garment, glowing gold eyes (job d01d51f9). Hard-won lesson logged: the marble-statue MEDIUM is fixed for everyone, only attire/features/attributes vary; a prior attempt that made her a realistic/cartoon person was rejected. Sheet: CLAUDE OUTPUTS/Affirmology/Affirmology_Sol_and_FinalVoices_v1.html. (For an exact video-grade likeness, train a Higgsfield Soul from more Sol photos later.) OSHUN = LOCKED (2026-06-18) as a WATER ELEMENTAL: her West African face surfacing and taking marble form out of a churning emerald-and-gold river, water all around, gold beaded crown + collar, glowing gold eyes, half-emerald/half-cream marble (job e5a1a470; takes 76247d60/bc056c62/09fd9b52 are the alts; the earlier honey/clear-water versions are superseded). She takes the West African seat (covers griot/orisha). HILDEGARD = LOCKED as the Christian mystic-healer (veil, vision-wheel, healing herbs, job eda30b81). FEMALE TAO = LOCKED (screen-facing, flowing water/yin-yang/jade, job c6a448b9; covers I Ching + Quan Yin). PARKED ideas still without a bust: a Dogon star-sage, Laozi, a Kabbalist (Jewish Tree of Life, likely just the Tree motif + Wizard Hermes covers it), an Eastern Orthodox light-mystic, an Aboriginal Dreamtime voice, a Polynesian sound-healer priestess (in progress), optionally Campbell/Wilber. Sheet: CLAUDE OUTPUTS/Affirmology/Affirmology_WisdomCircle_Sheet_v1.html. (GAIA was also reworked as an Indigenous earth-mother woman, job 176f2987, handled respectfully with no specific sacred regalia.) SOPHIA = lead oracle, flowing waved hair, gold zodiac diadem, constellation threads to a heart-star, cream marble. Both saved as REUSABLE HIGGSFIELD ELEMENTS named "Atlas" (id 17c8ddc8-5684-42e5-901e-5842349a4b6d) and "Sophia" (id 107e9d5a-1d5f-4b73-b7ba-3781659577f2). PROMETHEUS = young Black man, dark bronze marble, a molten fire-heart cut-out blazing in the chest, a crown of flame from the head (job b92285c7). HYPNOS = hooded verde-green marble, a gold dream-mask over the eyes with glowing eye-slits, gold poppy (job 39062581). WHOLE ROSTER busts generated (glow state) and many refined to Jeff's notes: Athena (owl, tight updo, cream), Pythagoras (tetractys), Orpheus (lyre, honey), Apollo (mature, sun-disc, cream), Chiron (LOCKED 2026-06-18 as a Middle Eastern Jewish Essene healer-mystic, olive-bronze marble, linen head-wrap, healer's staff with herb, gold kintsugi scar; reworked so he reads distinct from Apollo and adds Levantine representation, job 9a0faaa2), Persephone (black-and-gold Nero), Echo (verde sound-rings), Hermes (the warm "baller" messenger, refined to a bigger caduceus WAND + small wings on a worn back-strap, no silly hat; LOCKED on the refined 2nd take, job 0b609b93), Eros (kept at TWO ages, younger and ~thirty), Concordia (LOCKED as a Black African woman holding glowing gold scales, job 5a7b9ef2; she takes the African-woman seat, so Gaia stays bronze), Hestia (older grandmotherly), Gaia (bronze, world held). HERMES SHAPE-SHIFTER FORMS generated (he morphs across worlds): Greek base (locked), Thoth (Egyptian ibis-scribe, job 86f52d31), Quetzalcoatl (feathered serpent, 2b09f14e), Hermes Trismegistus (alchemist-sage, 1f8bc9fd), Atlantean / Emerald Tablets mystic-king (14d3e2b7, deliberately NOT ibis-faced). SRI AGASTYA replaces Parashara as the Vedic/Nadi seat (bronze, palm-leaf scroll). ARCHITECTURE: the cast is TWO CIRCLES, a small fixed Core Engine (Sophia, Sri Agastya, Prometheus, Athena, Pythagoras) plus an expandable WISDOM CIRCLE of world-tradition voices (Rumi locked male; Siddhartha; Lakshmi; Isis; Ix Chel for Mayan; Quan Yin; Oshun + a Dogon star-sage for Africa; Celtic Taliesin/Cerridwen; a Siberian shaman; animism/spirit-totems as a future upgrade). Doc: CLAUDE OUTPUTS/Affirmology/Affirmology_CouncilArchitecture_WisdomCircle_v1.md. WORLDS (backplates) generated: a locked emerald-and-gold COUNCIL CHAMBER (domed oculus, central orrery, above the clouds), a grand ancient TEMPLE-LIBRARY/observatory with an open center, a SKY-TERRACE overlook, and a serene FEMININE SANCTUARY (two variants, no-bath and background-bath). DELIVERABLES (all in CLAUDE OUTPUTS/Affirmology/): Affirmology_OracleHeroSpec_Sophia-Hermes_v1.md, Affirmology_CharacterSheet_BuildStandard_v1.md (the full-sheet standard, sourced from AI UPGRADE/Knowledge/Video-and-Characters/), Affirmology_HeroCharacterSheets_Sophia-Atlas_v1.html, Affirmology_CouncilLookbook_v1.html, Affirmology_VisualGallery_v1.html, Affirmology_CouncilBustSheet_v1.html, Affirmology_Explorations_Prometheus-Hypnos_v1.html. (Note: the chat image previews were failing on Jeff's end, so all images are delivered via these local HTML galleries that load from Higgsfield's CDN.) PIPELINE / NEXT: Nano Banana Pro busts -> reusable Higgsfield Elements/Souls (train a Soul from 10-25 views to lock identity for video) -> full character MODEL SHEETS (full-body turnaround front/3-4/side/back + expression strip + key poses + face/attribute callouts + color swatches, one canvas, eyes always glow or amber) -> Seedance 2.0 video. NEXT STEPS: finalize Hermes; train Souls for the locked heroes; build full model sheets; then generate the Wisdom Circle busts. Higgsfield is on the Plus plan (1,200 credits/mo, ~2 credits per Nano Banana Pro image); the false "out of credits" wall earlier was a concurrency misfire, plenty of budget remains.
  • Colin advisor report delivered for sending: CLAUDE OUTPUTS/Affirmology/Affirmology_FoundingChart_forColin_v1.pdf (Jeff+Sol base, then Colin woven in as advisor; readable type pass applied).

  PREMIUM REPORT ENGINE v10 (2026-06-17, Cowork lane) - shared between both lanes

  Built a self-contained reportlab report engine that finally embeds the REAL brand fonts (Cormorant Garamond + Inter + JetBrains Mono, instanced from Google Fonts variable files) instead of the Liberation substitutes the old builds used. This is the single biggest premium lift the redesign plan flagged. It lives in CLAUDE OUTPUTS/Affirmology/report-engine/ so Claude Code can use it too (Jeff wants tools fluid between lanes): report_engine.py (reusable builder: emerald cinematic page frame + starfield, full-cream ~11.5pt body, gold mantras with flanking rules, oracle-voice callout cards, convergence cards, founder panels, constellation/diamond motifs + a hand-drawn cover), assemble_report.py (style-aware extractor that reconstructs an existing report PDF into ordered role-tagged blocks, text verbatim, so NO content drift), build_jeffsol_v10.py (config), jeffsol_content.json, and fonts/static/*.ttf. Regenerated Jeff & Sol as Affirmology_CosmicBlueprint_Report_v10.pdf (16pp, montage-reviewed, cover is a clear upgrade). The locked APPROVED FORMAT (continuous flow, panels, oracle ORACLES, sparing gold mantras, overview, "What This Asks of You", future-journal close) is preserved exactly; only the typography/beauty changed. Chart-data panel font bumped slightly per Jeff. Jeff's design notes (2026-06-17): the look is good, the small chart-data font was bumped, investor/business docs do NOT need the emerald background (white is fine for those), and real graphics come LATER from a visual-aids suite (current motifs are light placeholders). NEXT: apply the engine to Colin (v10) and make it the template for ALL future reports; backport into the Mac builders or just use this engine. The investor/exec standalone PDFs (FriendsAndFamily, InvestorBrief, ExecSummary, data-room one-pagers, North Star) are DRAFTS and LOWER priority per Jeff; if/when done, correct the ExecSummary's stale "distributions from day one" framing to the locked v2 capital philosophy, and they can be on white, not emerald.

  Report format LOCKED (2026-06-16) + two reports final

  After several iterations, the report format Jeff approved is: continuous flow (no forced page-per-section), compact "mechanical" panels with RICH explanatory + cross-system SYNTHESIS prose underneath, the named agents as "ORACLES" giving short signed blurbs throughout plus a longer Oracle council conversation, a few sparing gold mantras, an OVERVIEW up top, a "What This Asks of You" ACTIONABLE section, and a future-journal close. Build with reportlab + embedded Liberation fonts; ALWAYS run a render-every-page montage review for deadspace before delivering. Builders: outputs/build_js_v9.py (Jeff & Sol) and outputs/build_colin_v3.py (Colin). Finals: CLAUDE OUTPUTS/Affirmology/Affirmology_CosmicBlueprint_Report_v9.pdf and Affirmology_FoundingChart_forColin_v3.pdf. This is the template for all future reports (the Reporting System Overhaul project). Sol correction baked in: she is the orchid/voice/pulse who shines when invited; Jeff convenes community in person and sets her stage.

  15. Kickoff-week audios rendered via the Studio on the Mac (2026-06-17)

  Correct path locked in: this chat drives the Studio API over the Tailscale funnel (https://macbook-pro-2.tail333640.ts.net:8443, token in affirmology-studio/.studio_token). compose -> audio -> poll jobs; the render runs ON THE MAC (chart-driven, council loop, QC-gated, Heavenly Circuit bed). NO in-sandbox synthesis, NO chunking. The earlier failure was trying to synthesize inside the Cowork sandbox (hard 45s command cap) during an ElevenLabs slow patch; not a token/credit/server problem (account healthy, ~75% chars left). Rendered + delivered to CLAUDE OUTPUTS/Affirmology/KickoffWeek_Audio/: - Sol_TheCrowning_v1 (astro_moment, sol, 8:14) - Jeff_RePourAndOverflow_v1 (astro_moment, jeff, 7:33) - Wealth_Is_The_Weather_Jeff_v1 (before_bed, jeff, 7:50) OPEN: Solstice Launch is a TWO-PERSON team audio; Studio compose is single-person and there is no team/relationship category wired in yet (relationship.py exists in the engine but is not exposed via the Studio). Needs a decision: single-person solstice render now, or wire the team path into the Studio (with Jeff). Also: team-script Sol-role fix applied in Affirmology_KickoffWeekSuite_v1.md (02/06); combined Jeff "Re-Pour and the Overflow" added as Script 09.

  16. Companion (two-person) audio: design + Model A delivered (2026-06-17)

  Spec: CLAUDE OUTPUTS/Affirmology/Affirmology_CompanionAudio_TwoVoice_Spec_v1.md. Decision (Jeff): ship MODEL A now (one narrator, "we" + occasional direct address to each); flagship later is Model B two-voice with Jeff/Sol CLONED voices (instant clone ~1-3 min calm sample each; pro clone 30+ min). Per-person voice_id to live on the Studio profile. Delivered the 4th kickoff audio via brief-driven compose on the existing single-person path (anchor jeff + astro_moment + relationship facts in the brief): SolsticeLaunch_Team_v1 (6:09). Came out as a true "we" piece with shared Cancer Moon / Scorpio Mars body integration + direct address to each. NOTE: engine named Jeff's Pearl as Gate 45 Synergy (his real chart) vs the brief's 44 teamwork; minor, chart-grounded. DURABLE BUILD (next, with Jeff): proper team path = generalize relationship call to N people, team composer with speaker tags [JEFF]/[SOL]/[BOTH], Studio "Companion" category with people[] on /api/compose. start_public.sh runs uvicorn WITHOUT --reload, so adding the endpoint needs a server restart by Jeff to go live.

  17a. Music toolkit + strategy (2026-06-17, Cowork)

  Catalog of music sources with license terms: Affirmology_MusicToolkit_Catalog_v1.md (uploaded). Audition pad: Affirmology_MusicAudition_Log_v1.md. STRATEGY DECISION (Jeff): cheap route to proof-of-concept and FUNDING first, custom beds later. Phase 1 (now): Suno (already subscribed, $0 incremental) for on-demand beds + 2-3 pay-once Meditation Music Library beds only where the clean sold-product license matters (sleep/journey/morning); no subscriptions or buyouts yet. Phase 2 (after funded): submit the Sonic Spheres quote form (https://forms.gle/7xkDpx2NvEjRcyKL6) and commission owned, on-brand signature beds per use case (the Affirmology house sound); revisit TunePocket/PremiumBeat if still useful. Sonic Spheres has NO public pricing (quote-only). LYRIA 3 (Google) added to the toolkit (2026-06-17): Lyria 3 Fast ~30s, Lyria 3 Pro up to ~3min structured, commercial-OK on paid Gemini tiers, SynthID watermark, on Gemini API/Vertex (~$0.08/song) so programmable; best NOW for marketing soundtracks (pairs with Higgsfield), candidate for product beds, later API per-user music (verify Vertex redistribution-at-scale first). Consolidated index of ALL music tools + links: Affirmology_MusicTools_MasterList_v1.md; licensing detail in Affirmology_MusicLicensing_v1.md. The locked demo bed (Heavenly Circuit) is untouched by any of this. SUNO BED PROMPT PACK built (2026-06-17): Affirmology_SunoBedPrompts_v1.md, ready-to-paste instrumental/loopable bed prompts (3 variants each) for Sleep, Journey, Morning, Gym, with an Affirmology sonic signature; Jeff generates in his Suno Pro/Premier account (commercial-cleared) and logs keepers in the audition log.

  17c. Voice tools: Fish Audio evaluation (2026-06-17, Cowork)

  Deep multi-source eval of Fish Audio (OpenAudio / Fish Speech, flagship S2 Pro) vs ElevenLabs for two Jeff use cases: (1) real-time "talk to Sophia" avatar installation, (2) cloning a person's voice into the sold Sacred Audios. Doc: Affirmology_VoiceTools_FishAudio_Eval_v1.md. Findings: Fish S2 rivals/beats ElevenLabs on raw naturalness (vendor-run blind test; independent leaderboards put it mid-pack), far cheaper (~$1.25/hr audio), great breathy/calm emotion tags. KEY TRAPS: Fish open weights are NON-COMMERCIAL (Fish Audio Research License + CC-BY-NC-SA; the "MIT/Apache" claim online is stale, verified via raw LICENSE), so self-hosting for a SOLD product is a license violation, the license-clean path is Fish's PAID hosted API. Fish has NO consent-verification (ElevenLabs enforces a voice-captcha); neither indemnifies you. RECOMMENDATION: keep ElevenLabs as production voice; ElevenLabs PVC (not Instant, not Turbo; stability ~35-40%, similarity ≤80%) is the more robust path for long-slow Sacred Audio AND the real fix for the Turbo dropout issue. Fish is the stronger fit for the avatar-installation voice (HeyGen LiveAvatar accepts Fish as TTS; only top option self-hostable offline for a kiosk) and worth a cheap A/B (~$11 Plus plan) on one real script. Legal: cloning a real person into a sold product needs a signed consent+commercial-use release per person + AI-generated disclosure to listeners (EU AI Act Art 50 from 2026-08-02; TN ELVIS Act; NO FAKES not yet law). Locked demo unchanged.

  17b. Atlas app (2026-06-17; DECISIONS LOCKED 2026-06-18, Cowork)

  Cloud move is DONE (section 19: Studio live on Render + R2), so Atlas is unblocked. Spec: Affirmology_Atlas_Overview_v1.md (sections 10-14 hold the locked decisions; builds on the thin Affirmology_Atlas_AppSpec_v1.md). Concept (locked): Atlas = the app, Hermes = the in-app assistant; two modes (Hermes chat default + full Studio one tap away) with a persistent corner toggle; installable PWA on phone + Mac; one Cloudflare Access login. DECISIONS LOCKED 2026-06-18 (Jeff): - Messaging circle v1 = Jeff, Sol, Colin only; messaging is IN-APP ONLY for v1 (email/external reach deferred). - Hermes TAKES ACTIONS (not read-only): creates audios from concepts, runs the composer, logs feedback, stores/pulls info wherever relevant ("the messenger"). - PERMISSION TIERS: Jeff = owner (everything, incl. private business material). Sol + Colin = collaborators: create/iterate/render audios, log feedback (text+voice), ask anything, message the team, read the shareable knowledge base and the investor-pitch/research/product material (they may be dropping insights into exactly that). They CANNOT overwrite/delete permanent files, and CANNOT see private ownership material (cap table/shares, Jeff+Sol founder agreements, legal terms). Future people: narrower. Reason Atlas matters: it is how Sol/Colin use the whole system without living in GitHub/Claude/Gemini. - Default landing = REMEMBER LAST MODE per person, with the always-present toggle. - Voicenotes = TRANSCRIPT ONLY (do not retain raw audio in R2). - Nightly improvement loop = AUTO-APPLY SAFE CHANGES + report the rest. "Safe" fenced hard: only low-risk reversible changes (copy/content/config/triage/draft proposals); NEVER the locked demo, NEVER delete/overwrite permanent files, NEVER auto-deploy engine/chart/script-structure code, NEVER silently spend voice credits; anything touching audio content/charts/structure/money/demo is report-and-propose for human approval (recommend git with easy revert). - VOICE TRANSCRIPTION (CORRECTED 2026-06-18): preferred path is NVIDIA Parakeet ON-DEVICE via sherpa-onnx (exported ONNX run on the phone CPU, offline, NO GPU; prebuilt iOS/Android examples exist for Parakeet-TDT-0.6b). This is what Jeff's dev demoed on a non-NVIDIA phone, and it is genuinely better: offline, zero per-use cost, private (audio never leaves the phone, which fits "transcript only" further), fast. Tradeoff: nudges Atlas toward a thin native wrapper or sherpa-onnx's WASM-in-browser build vs a pure PWA; settle at B7. If the dev's phone module works, route through it and skip server transcription. FALLBACK only: server-side OpenAI Whisper API on Render (seamless cloud call) to get something live before the native piece is ready. ("Wispr Flow" = Jeff's personal dictation app, not the app's layer; Granola = meeting notetaker, not for this.) - CLOUDFLARE ACCESS LOGIN is off-brand; brand the Access app (emerald/gold/cream + wordmark) or use a branded splash that hands off to Access. OPEN MEDIA NOTE: R2 public serving for media.affirmology.ai was 403 (section 19); Jeff believes it was fixed/tested OK overnight 2026-06-17/18 - VERIFY before relying on it.

  ATLAS v1 BUILD STARTED (2026-06-18, Cowork, building DIRECTLY in affirmology-studio since Jeff asked Cowork to build it; Atlas == the Studio repo + PWA + Hermes, not a separate app). All ADDITIVE, existing Studio pages/flows untouched, static-checked (node --check, python ast). UNCOMMITTED in the working copy; deploys to studio.affirmology.ai on commit+push (autoDeploy). Built so far: - PWA layer: web/manifest.webmanifest, web/sw.js (conservative: caches only the static shell, never API/media/auth/POST/cross-origin, network-first), brand icons web/icons/ (gold Cormorant "A" on emerald, 192/512/180/32), apple-touch + manifest + apple-mobile-web-app meta + SW registration in index.html. Installable on iPhone/Android home screen + Mac dock. iPhone Add-to-Home-Screen works on the manifest+meta alone. - ATLAS TWO-MODE SHELL: a fixed corner toggle (#modeToggle, bottom-left; shifts right of the sidebar on desktop) flips between HERMES (chat) and STUDIO (workbench); remembers the last mode per person (localStorage atlas_mode). New #page-hermes chat screen reusing the brand chat styles. - HERMES HOME HERO (2026-06-18, Jeff asked): Hermes is now the FIRST-CHOICE entry on the Studio Home screen, a prominent constellation-dripped card at the top (.hermes-hero in index.html): a twinkling SVG gold constellation (connected stars + scattered dim stars, soft glow filter, prefers-reduced-motion respected), "Talk to Hermes" headline, "Enter Hermes ->" CTA, taps straight into chat mode. APP ICON upgraded to a sexier cinematic version (glowing gold gradient Cormorant "A" with bloom, soft halo, faint starfield, sparkle, whisper of a gold ring) at all sizes in web/icons/. - HERMES backend POST /api/hermes (api/main.py): permission-aware Claude call. Owner (jeff) = full access incl. private business; collaborators (sol/colin/josh) = product/system/research yes, private equity/legal/agreements withheld. System prompt carries the brand rules (no em dashes, Sol not Soul, Subconscious OS, recognition over instruction, no meditation-teacher name, Tier C wall) + the four charts + curated context. Degrades gracefully if no key/model. Uses ANTHROPIC_API_KEY (already in the studio env for script gen); model via env HERMES_MODEL (default claude-sonnet-4-6). Front-end Hermes chat persists history per person in localStorage. - ALSO FIXED the S3 backend half: compose minutes clamp widened 20 -> 60 in api/main.py so the new 3-45 min length slider is actually honored by the engine (was silently capping at 20). STILL TO BUILD on Atlas: in-app (Hermes-side) messaging between Jeff/Sol/Colin, the approvals queue, voicenote capture (server-side Whisper interim; on-device Parakeet later), deeper Hermes actions (kick off a render, file structured feedback) + real knowledge-base/doc retrieval, and Cloudflare Access page branding. NEEDS ON DEPLOY: confirm ANTHROPIC_API_KEY present on Render (it is, for script gen) and optionally set HERMES_MODEL. Mobile studio nav is still sidebar-only (hidden on phones) - the mode toggle works on mobile, but studio page nav on mobile is a follow-up.

  ATLAS UPDATE (2026-06-18 later): (1) PUSHED (commit 6670b79): Hermes persona (humor calibration so he is not funny all the time + app-support scope + routing to the oracle specialists + a "how Atlas works" guide) and the in-app messaging BACKEND (messages table + /api/messages). (2) MESSAGING DEFERRED: Jeff is unsure it is needed; the backend rides along dormant/harmless, no UI built (task parked). (3) NEW + UNCOMMITTED (needs a push): CROSS-DEVICE HERMES HISTORY. Hermes turns now persist server-side per person (hermes_history table + add_hermes_turn/list_hermes_history in jobs.py; /api/hermes saves both turns and builds context from the stored history; new GET /api/hermes/history; front-end initHermes loads it). So a chat follows the person computer->phone (Jeff noticed history did not carry over). Files: api/jobs.py, api/main.py, web/app.js, web/index.html (app.js bumped ?v=17). (4) DONE (same uncommitted batch): VOICENOTES v1 = speak-to-text via the browser Web Speech API (transcript-only, no audio stored, feature-detected/hides if unsupported). Mic buttons in ALL boxes: Hermes input, composer input, composer notes, per-version feedback (#fbInput), and the global feedback popup (#fbText). toggleDictation()/speechSupported()/hideUnsupportedMics() in app.js; .micbtn styles + buttons in index.html. CAVEAT: Web Speech API can be flaky in an iOS standalone PWA; upgrades if so = server-side Whisper (needs a key) or on-device Parakeet via the native wrapper. The iPhone keyboard's own dictation mic also works in every field as a fallback. (5) JEFF PREFERS A NATIVE APP FEEL over the web app; the thin native Capacitor wrapper (cheap, store-free via TestFlight/sideload, also unlocks on-device Parakeet voice) is teed up for when he wants it.

  HERMES EFFECTIVENESS - PHASE 1 BUILT (2026-06-18, Cowork; UNCOMMITTED, needs a push + LIVE TEST). Trigger: Sol asked Hermes for her+Jeff "wealth codes" across Gene Keys/HD/astrology/full-numerology + 2026 transits and got useless results. Root cause: Hermes was a single Claude call with a THIN static prompt (only ~5 placements/person, no engine/corpus/oracle access). NOT an Atlas data gap (the engine has full charts) but a gap in the initial Hermes build. Fix: HERMES IS NOW AGENTIC (tool use). New api/hermes_tools.py + a tool loop in /api/hermes: tools = get_person_chart + compute_chart_from_birth (full astrology/HD/all-3-GeneKeys-incl-Pearl/full-numerology/Vedic via render_all_facts, engine already in cloud), search_corpus (the ~17k records), consult_oracle (sophia/athena/prometheus/pythagoras/parashara specialist sub-readings). System prompt updated to USE the tools. Cost-bounded (max 6 rounds). Files: api/hermes_tools.py (new), api/main.py. Plan + full Phase-2 list: CLAUDE OUTPUTS/Affirmology/Affirmology_Hermes_Effectiveness_Plan_v1.md. PHASE 2 PENDING: full Vedic engine (today only partial sidereal; collect Vedic for Jeff/Sol), Transit & Timing agent (Venus/Jupiter/Pluto 2026), store confirmed full LEGAL names for name-numerology, the KNOWLEDGE BRAIN (sync deep docs hard-drive->R2 via nightly cron + search_docs), the AGENTIC COUNCIL (oracles talking to each other; wire Chiron+Persephone), new client accounts in R2 + client tier. ROSTER NOTE: roles corrected (Sophia=Western astro+synthesis lead, Athena=Gene Keys, Prometheus=Human Design, Parashara=Vedic, Pythagoras=Numerology, Hermes=assistant/router).

  CRAFT KNOWLEDGE ENGINE (2026-06-20, Jeff flagged as the next pillar). Plan: Affirmology_CraftKnowledgeEngine_Plan_v1.md. GAP (verified in code): the cosmic blueprint got the full treatment, but the CRAFT side (neuroscience, frequency/sound, meditation/hypnosis technique, audio production, pro reading craft) is shallow, just a STATIC TECHNIQUES_LIBRARY menu in agents/build_report.py that Chiron picks from; the deep research sits in ~10 docs, not distilled/actionable/growing; the audio mix does NO frequency/binaural work (just a music bed). UPGRADE = a second knowledge pillar (blueprint = WHAT we say, craft = HOW we make it land): a "craft" corpus/meaning tradition seeded by distilling our research docs into actionable entries, wired DEEPLY into Chiron + Orpheus + Apollo's efficacy gate + the audio mix + Hermes (search_craft tool, replacing the static menu), and grown by the nightly cron (crawl wide). Optional later: binaural/solfeggio/tonal overlays in audio_mix.py (the one piece that changes the actual SOUND; Studio-only, demo locked). Code prompt + phased build in the plan doc. HERMES LOOP BUG fixed same day (see below / Affirmology_HermesLoop_Bug_v1.md); needs the api/main.py+hermes_tools.py push + local restart.

  TEST & CHANGE QUEUE (2026-06-20) - the single living list of pending tests + changes, prioritized P0-P3, each with the exact next action: Affirmology_TestAndChange_Queue_v1.md. This is the operational hub; ADD every new "we should test/change X" here, and check items off as done. A daily 8am scheduled digest ("affirmology-test-change-digest") surfaces the top P0/P1 items to Jeff. Craft-engine agent ownership corrected: HYPNOS is primary (subconscious-efficacy + frequency/entrainment), with the Neuroscientist (the science), Orpheus (writing craft), Apollo (the gate), Chiron (healing slice only), Echo (feedback loop), not Chiron-as-primary. Frequency/sound layer (binaural/solfeggio overlays in audio_mix) = Jeff WANTS it; filed as queue C3, to spec.

  COSMIC BLUEPRINT ENGINE UPGRADE (2026-06-18, Cowork) - "be masters of the cosmic blueprint." Master gap analysis (from another chat): CLAUDE OUTPUTS/Affirmology/Affirmology_EngineCoverage_Gaps_v1.md (7 priorities). HOW-to-build brief: Affirmology_CosmicBlueprint_EngineUpgrade_BuildBrief_v1.md. BIGGEST gap = the engine computes NO ASPECTS (positions without aspects = a list, not a reading). BUILT (Cowork, additive new modules, UNWIRED, need integration+test on the Mac because they touch the locked demo chart path): affirmology-agent/src/affirmology/chart/aspects.py (aspect grid: major+minor, orbs, applying/separating) and chart/extended.py (no-data-file: Black Moon Lilith mean+true, Part of Fortune/Spirit sect-correct, Vertex/anti-Vertex; data-file: Chiron + Ceres/Pallas/Juno/Vesta best-effort). NOTE: the no-file points were already computed ad hoc for Jeff & Sol by the other chat (out/*/extended_points_facts.md) = a regression anchor. PENDING (Code/Mac, sequence with corpus session; all chart/+reports/+Dockerfile, low collision): wire both modules into compute_astrology + models.AstrologyChart + reports/render.py render_all_facts (so it flows to reports AND Hermes), install seas_18.se1 in api/Dockerfile (+ AFFIRMOLOGY_EPHE_PATH, switch asteroids to FLG_SWIEPH), pytest + demo-chart regression, re-vendor/deploy. THEN Vedic depth (Vimshottari dasha, vargas D9/D10/D2, dhana yogas), condition/timing + transit-to-natal (also powers Hermes timing), HD/numerology/Gene-Keys depth (Priorities 3-7).

  DOMAIN PLAYBOOKS (2026-06-18, Cowork) - Jeff: the system should already KNOW how to use the blueprint for predictable life questions (money, relationships, career) so agents don't figure it out on the fly. Model: tradition oracles read by SYSTEM; DOMAIN agents read by QUESTION (which cross-system signatures matter + how to synthesize), plugging into the agentic consult_oracle. Built: CLAUDE OUTPUTS/Affirmology/Affirmology_DomainPlaybooks_Lakshmi_v1.md = LAKSHMI's full wealth playbook (she is already the Wisdom Circle abundance voice per Affirmology_CouncilArchitecture_WisdomCircle_v1.md) + the life-domain->voice map. Wealth signatures she reads: Gene Keys Pearl sequence, astro 2nd/8th/11th + Jupiter/Venus/Part of Fortune/MC, Vedic Dhana yogas + D2 Hora + dasha timing, numerology 8/4/22, HD type/Heart-Ego/Sacral/not-self. DECISIONS (Jeff 2026-06-18 late): CAREER/vocation = ATHENA ("unless/until someone better"); HEALTH/body = IX CHEL (decided); QUAN YIN removed (FEMALE TAO absorbs compassion/being-heard); Christian mystic = HILDEGARD; AGASTYA confirmed (Vedic seat). COUNCIL WIDENED (Jeff: "be smart, not too limited"). The agentic consult_oracle (api/hermes_tools.py ORACLES) now has 12 voices: tradition experts sophia, athena (+career/vocation), prometheus, pythagoras, AGASTYA (renamed from parashara, the Vedic seat); client-facing DOMAIN voices lakshmi (wealth), eros (love), concordia (partnership/teams), hestia (family/home), gaia (community/foundation), persephone (transformation/breakthrough), chiron (healing). The domain voices can speak directly to clients. HERMES himself can also speak directly, especially on ALCHEMY (his Hermes Trismegistus alchemist form), wired into his persona prompt. More can be added over time; deeper/obscure Wisdom voices stay premium.

  READING CRAFT STANDARD (the "Forrest standard", from another chat's CLAUDE OUTPUTS/Affirmology/Affirmology_ReadingCraft_Standard_v1.md). Diagnosis (correct): strong DATA layer, thin INTERPRETATION layer, so the agents grab a keyword ("8th house = other people's money") instead of teaching the MECHANISM + the choice. The 7 disciplines: mechanism-not-label, free-will-not-fate, high-road/low-road, synthesize, one living metaphor, talk warmly, end on a choice. APPLIED NOW (Cowork, uncommitted): a CRAFT block folded into all 12 Hermes consult_oracle oracle prompts + Hermes's synthesis (api/hermes_tools.py, api/main.py), so readings improve at the source. STRATEGIC (the "could be big" lever): the DEEP MEANING LAYER + DISTINCTIONS LIBRARY should be a structured store the NIGHTLY CORPUS IMPROVER builds + grows, and every Jeff/Sol "this reads odd" feedback becomes a permanent distinctions entry (corrections loop = feedback store -> nightly improver -> distinctions). The CRAFT-CRITIQUE PASS (an Apollo-style QA gate that rejects determinism/keyword-dumping/missing-high-low-road/house errors) is the reliability lever, build FIRST. Aspects (Phase 2) are the prerequisite for the synthesis discipline. Sophia already is the humanistic-Oracle voice. Recommended order: prompt-craft (done) -> critique gate -> meaning layer + distinctions via the nightly improver.

  STRUCTURES ENRICHED with the new blueprint (2026-06-18, Cowork, UNCOMMITTED, rides the Hermes-layer push). api/structures.py: a global BLUEPRINT_ENRICHMENT directive is appended to EVERY audio's generation in api/main.py compose (so all audio types, incl. the natal/demo structure, now draw on the aspect grid + Lots + Lilith + Vertex + Chiron + Vedic, read with craft, using only what serves the piece). Plus targeted per-structure directive tweaks: gk_pearl (Fortune/Spirit), gk_venus (Venus aspects + Vertex), confidence_abundance (Lilith/Fortune), before_bed (Chiron/Lilith gently), founder (Spirit + MC aspects), astro_chart_power + all_systems_agree (the aspect grid). api/overview.py appends a "DEEPER READ" note to every structure's compose panel. NOTE: this enriches STUDIO generation; the separate LOCKED demo render server would need the same enrichment wired to get aspects into the live demo audio (coordination item). ORCHESTRATION (dynamic, not fixed): usually the core astrologers (Sophia/Athena/Prometheus) + Hermes bear the message with specialists feeding support; sometimes the bearer asks a specialist to go deep; sometimes a domain voice (Lakshmi/Eros) speaks to the client directly. UNCOMMITTED: hermes_tools.py + main.py (Lakshmi + Eros oracles) need a push. Next playbook: relationships (Eros live + the lenses).

  17c. Voice engine alternative to test: Fish Audio (2026-06-17, Cowork)

  Jeff's call: start creating tests with Fish Audio (fish.audio, by Hanabi AI) as an alternative voice engine to ElevenLabs. Evaluated 2026-06-17. What it is: an expressive real-time TTS + voice-cloning platform (current model Fish Audio S2; open-source lineage OpenAudio S1 / Fish Speech), with a hosted API and an open-source self-host path. Why it is worth testing for us: - Cost. API is roughly $15 per 1M UTF-8 bytes (s2-pro), pay-as-you-go, no subscription or monthly minimum. Independent and Fish's own comparisons put it about 50 to 70 percent cheaper than ElevenLabs for comparable quality. At our render volume this matters once the Studio scales. - Voice cloning from a very short sample (they claim 10 to 15 seconds), multilingual (30+ languages, same voice). Directly relevant to the Model B flagship two-voice plan (cloned Jeff and Sol voices, section 16): a cheaper, faster clone path to A/B against ElevenLabs. - Inline emotion and delivery tags ([soft], [whispering], [breathy], [sighing], [pause], [long pause], [excited], etc.). Promising for our meditative cadence and for letting key recognitions breathe, which is an open Studio craft item (section 13). - Open-source self-host option for cost and control at scale, and an API with low-latency streaming for the future Atlas voice-note / chatbot directions. Hard constraints on any test (do not break): - THE DEMO STAYS LOCKED to ElevenLabs Charlotte at the proven baseline (section 4). Fish is a STUDIO / sandbox experiment only. No Fish output goes near the demo unless Jeff explicitly says "make this the new demo." - scripts/audio_qc.py still gates every Fish render (dropout/artifact reject), same as ElevenLabs. Verify Fish honors a reliable, warble-free pace before trusting it; our whole baseline lock exists because Turbo warbled at speed. - Commercial use needs a paid Fish plan (their free tier is personal-use only). Confirm commercial rights and data/privacy terms before any client-facing use, especially for cloned founder voices. - Voice identity: blind A/B Fish vs ElevenLabs for (a) a cloned Jeff/Sol voice and (b) a Charlotte-equivalent narrator, judged on authenticity, emotional nuance, and artifact-free long renders, before adopting anywhere. NEXT: stand up a small Fish test in the Studio sandbox: clone Jeff and Sol from short calm samples, render the same script through Fish and through ElevenLabs, QC both, and compare. Pricing/API source: docs.fish.audio + 2026 cost write-ups.

  17d. SMS vs email delivery for sample/free/affiliate audios (2026-06-17, Cowork)

  Jeff's question: automate SMS delivery instead of email for better response on sample/free/affiliate Affirmology audios; concern that email is needed for the Meta ads pixel. Looked at quo.com + 2026 benchmarks. Findings: - Response gap is real: 2026 benchmarks put SMS open ~90-98% vs email ~28%, SMS CTR ~19-22% vs email ~3% (about 9x). SMS wins the single high-intent "your audio is ready, tap to listen" moment. - The pixel premise is slightly off, and this changes the decision. The Meta pixel fires on the WEB PAGE, not inside the email or text. Whatever channel carries the link, the pixel fires when they land on the audio player page, so retargeting works either way. For audience matching, Meta Custom Audiences + Conversions API accept HASHED PHONE NUMBERS as a primary identifier, not just email. So SMS does NOT sacrifice Meta retargeting/audience building. Email's genuine edge is different: far cheaper per send, better long-form nurture, universal CRM key. - Therefore not either/or. Plan: capture BOTH email + phone on the free-audio form (phone with explicit SMS consent); deliver the audio LINK by SMS for response rate; the link lands on a pixel-tracked player page so tracking/retargeting works; keep email for cheap ongoing nurture + as a second match key. Audio is never attached; always a link to a hosted player page (where pixel + analytics live). - Real cost of SMS is COMPLIANCE, not the pixel: US marketing texts need explicit opt-in, 10DLC A2P brand+campaign registration, STOP/HELP handling, TCPA (statutory damages per message). Highest risk with AFFILIATE traffic: consent must be captured on OUR form with clear SMS-consent language, never assumed from an affiliate list. This is the gating workstream before scaling SMS. Tooling tiers: - Quo (formerly OpenPhone): a shared business phone + inbox for two-way conversations (calls, texts, AI receptionist, routing), NOT a bulk-marketing blaster. Has an API, Zapier/Make, a Quo MCP, and a Claude integration "coming soon"; automated texts via Zapier ~$0.01/msg. FIT: the personal/relationship layer (a real number that texts the audio link and can reply, for Jeff's Miami community + founder-led/affiliate concierge warmth). The "Claude integration coming soon" is worth watching for our agent stack. - Twilio or a marketing-SMS platform: the automated, scaled sample/free/affiliate funnel. Our engine already produces per-person output, so a send can fire on render-complete. Twilio ~$0.0079/SMS + carrier fees. NET: pursue SMS as primary delivery for samples/affiliate audios, keep email alongside, treat consent/10DLC as the gating workstream. Quo = human/relational sends; Twilio (or similar) = automated funnel. Sources: dotdigital/omnisend 2026 benchmarks, quo.com, Twilio pricing.

  DELIVERY FLOW DECISION (2026-06-17, Jeff + Cowork): - Do NOT gate the free/sample/affiliate audio behind an app install. For top-of-funnel, friction kills conversion: a stranger's first taste must be a no-install WEB PLAYER PAGE (PWA, app-like, zero install), and that page is where the Meta pixel lives and fires. The Atlas app is the UPSELL / retention layer for converted + community users, not the front door. After the audio lands, soft-prompt "add to home screen / get the full app." - Flow: SMS/email link -> pixel-tracked page -> listen -> app prompt. One page (player + pixel together) or two (a quick tracked landing that fires an intent event, then the player) both work. START with the single player page; only split into two if we want a separate pre-audio "intent" event to optimize ads against. The link CAN route through a tracking page then into the app if a given flow calls for it, but the default sample flow stays web-player. - Twilio is the right automation tool (programmatic SMS firing on render-complete, cheap pay-as-you-go); the spend is small. The real cost of going SMS is consent + 10DLC registration, not the Twilio bill. Budget TIME there, especially for affiliate traffic. - Data capture: capture BOTH email + phone (phone = ~9x SMS response + delivery channel; email = cheap nurture + second Meta match key). Every required field costs completion rate, so this is an A/B question, not a default. MAIN OPEN DECISION (Jeff, to settle): (A) PHONE REQUIRED + EMAIL OPTIONAL, or (B) REQUIRE BOTH. Lean: if SMS is primary delivery, phone is the must-have; option B maximizes data + reach but at some completion-rate cost. Test both.

  POST-AUDIO MONETIZATION LAYER (2026-06-17, Jeff + Cowork): - The post-audio moment is the highest-intent window (they just felt something true about themselves). Make the offer there. Existing asset pointed at this: whatsinstore.html post-audio upsell reel (section 8); the "marketing move pop-up" is that, tightened into a clear offer + button. - TWO coordinated surfaces, not redundant: (1) ON-PAGE post-audio pop-up = the peak-moment ask; (2) SMS sequence = re-engage non-converters over the next few days with a small set of different angles/landing pages. Every SMS pushes to a pixel-tracked landing page, so each one also feeds retargeting audiences (ad engine keeps warming). - UX: LET IT LAND. The pop-up appears one beat AFTER the audio ends, not on top of the afterglow (same "let recognitions breathe" craft value as the audios). - Offer set: keep small + test. Candidates: full personalized audio / subscription, deeper cosmic report, affiliate invite, bundle. - CONSENT COPY TREATMENT (Jeff's call): on the capture form, "we'll TEXT YOU THE AUDIO" in BIGGER text; the marketing-message consent in SMALLER, different-color text. Still clearly readable, just less of a vibe killer. COMPLIANCE NOTE: that styling is fine as long as the marketing disclosure stays legible and clear (not hidden/deceptive); delivering the audio is TRANSACTIONAL but pushing offers is MARKETING, so the opt-in must explicitly cover "we'll also text offers," not just the audio. Keep it clear-and-conspicuous to stay TCPA-safe (small ≠ buried).

  18. Film visual-asset pack (2026-06-17, Cowork) - the "make it not look lame" build

  All in CLAUDE OUTPUTS/Affirmology/film-assets/ (generators alongside, so anything is re-tweakable). The principle locked with Jeff: the browser is the COMPOSITOR/conductor, not the artist; structured charts I build procedurally (my strength), cinematic/organic footage comes from Higgsfield (Jeff, starting tomorrow). Guidance doc: Affirmology_FilmAssets_ShoppingList_v1.md. Process/diagnosis doc: Affirmology_CosmosFilm_EvolutionProcess_v1.md. The B2 cosmos film text-overlap was also fixed in Affirmology_Primer_B2_textfix.html (captions to lower third, sequential, no collisions; Claude Code to fold in).

  Built and Jeff-approved this session: - Zodiac: 12 gold constellation SVGs (zodiac/), elegant, drop-in. - Human Design bodygraph v1 (bodygraph.svg): correct 9-center shapes/positions + channels, centers id'd. NEXT (requested, not yet built): bodygraph v2 with CURVED channels, gate numbers inside centers, and the Design/Personality side panels; plus a flashing multi-chart beat (different pretend definitions glow back-to-back with type+authority words: Manifesting Generator/Generator/Manifestor/Projector/Reflector, Sacral/Emotional/Splenic authority, etc). - Gene Keys sequences (gk_seq/), shapes now CORRECT and locked (do not relitigate): ACTIVATION = zigzag (Life's Work top -> Evolution right -> Radiance left -> Purpose bottom). VENUS = zigzag chain Core->SQ->EQ->IQ->Attraction->Purpose (6 spheres). PEARL = tetrahedron (Brand/Pearl/Vocation/Culture). Each centered in the 64-gate wheel, color-coded (Activation emerald, Venus gold, Pearl pale). One-by-one light-up GIF exists (goldenpath/sequences_lightup.gif). - Full Golden Path (goldenpath/golden_path_full.svg): all 11 spheres, three sequences color-coded, with the SQ->Vocation link Jeff requested. Correct as of end of session. - Brain: brain_gold.png / brain_emerald.png (recolored from Jeff's brain2.png, transparent, glow). brain_alive.gif = breathing glow + firing synapses + absorption pulse (2D animation I make here; 3D rotation would be Higgsfield). - Sacred geometry (sacred/): Flower of Life, Seed of Life, Metatron's Cube, Merkaba, Sri Yantra, Golden Spiral, Torus, Vesica Piscis, all gold line-art; plus sacred_swirl.gif (counter-rotating hypnotic mandala). - Web feasibility demos (answer to "can we do quantumstretch / Codrops particle galaxies"): galaxy_demo.html (Three.js + GLSL animated galaxy, gold core to emerald arms, live sliders, manual orbit; FIXED the OrbitControls-404 bug) and scroll_demo.html (GSAP+Lenis preloader/smooth-scroll/pinned-reveal/counters in-brand). Both UNVERIFIED by Cowork (no browser/GPU in sandbox); Jeff reviews in browser. Verdict: yes, that whole class of site is our existing GSAP/Three stack. - STILL QUEUED (procedural, no Jeff assets needed): astrology natal wheel with lines drawing in + rotation (pairs with the birth-flash beat: Earth + horizon flare -> ring -> chart); bodygraph v2 + flashing charts (above). - TOMORROW: Jeff generates Higgsfield footage (cosmos nebula loop, Earth-with-sunrise, horizon flare per the shopping list); Cowork strips/recolors/composites it with these assets and can scroll-scrub it. A Higgsfield-style media MCP (generate_video/image/3d) also surfaced in this session's tool list.

  17. Demo overhaul: execution plan (2026-06-17)

  Consolidated, lane-tagged plan to resume the demo overhaul the moment Claude Code finishes the B2 cosmos loading film: CLAUDE OUTPUTS/Affirmology/Affirmology_DemoOverhaul_ExecutionPlan_v1.md. Covers the five flow pieces (pre-form opening, loading film, audio player, post-audio display, post-audio short video), the post-audio result.html bug notes (Safari scroll stall; thin blueprint missing Rising + more Gene Keys + intro; download buttons still point at mocks; consistency check), the email switch (Resend domain verified; still need RESEND_API_KEY + FROM_EMAIL on the worker), the ordered checklist, and a NEW proposal Jeff asked about: an AMBIENT (not kinetic) animation during listening, slow/low-contrast/optional, loosely audio-reactive, the user's chart barely visible behind it, with a dim toggle and prefers-reduced-motion. Jeff likes the idea; treat it as the sixth experience. - COSMOS FILM TEXT FIX (2026-06-17, Cowork): the B2 primer had captions colliding (all pinned dead-center with overlapping time windows, plus chart-art labels in the same zone). Cowork applied a deterministic fix in Affirmology_Primer_B2_textfix.html (copy of B2, original untouched): captions moved to the lower third (top:71%) so they clear the centered chart art, the three system label+phrase pairs merged into single two-line beats, and every caption window made strictly sequential (verified zero time-overlaps). CLAUDE CODE: fold these BEATS + the .beat position change into the canonical B2 on the next pass. Aesthetic/art quality is a SEPARATE problem (needs real assets, not code) tracked in Affirmology_CosmosFilm_EvolutionProcess_v1.md.

  19. CLOUD DEPLOY - Studio LIVE on Render + R2 (2026-06-18, Claude Code)

  The Studio engine now runs in the cloud, not just the laptop. Full render pipeline PROVEN end-to-end.

  • SERVICE: affirmology-studio-api on Render, LIVE at https://affirmology-studio-api.onrender.com (Docker from api/Dockerfile, blueprint render.yaml, autoDeploy on push to jeffparkerlove22/affirmology-studio main). Health: GET /api/health.
  • REPO: the studio is its own PRIVATE repo github.com/jeffparkerlove22/affirmology-studio. The engine (affirmology-agent) is VENDORED into it CODE-ONLY (rsync + .gitignore; no .env/.venv/out/music/db/media; proven with git check-ignore). To update the engine in the cloud, re-vendor + push.
  • STORAGE: R2 bucket affirmology-media. STUDIO_STORAGE=r2, R2_PUBLIC_BASE_URL=https://media.affirmology.ai. Rendered MP3 + PDF upload to R2 under <job_id>/.... R2 creds live in affirmology-agent/.env (gitignored) for the nightly upload job; same creds set as Render secrets.
  • CORPUS IN CLOUD: corpus.db (9.4MB) uploaded to R2 key corpus/corpus.db. On boot api/corpus_boot.ensure_corpus_db() pulls it to the persistent disk /data/corpus.db and sets AFFIRMOLOGY_CORPUS_DB. It RE-PULLS when the R2 object size differs from local (so a restart after a fresh re-upload loads the new corpus; no manual delete). GET /api/corpus-stats returns the live record count for verification. So cloud generation stays chart-grounded.
  • MUSIC BED: Heavenly Circuit.mp3 uploaded to R2 key music/Heavenly Circuit.mp3; api/corpus_boot.ensure_music_bed() pulls it to /data on boot (STUDIO_MUSIC_BED=/data/Heavenly Circuit.mp3). Media is never committed to the repo.
  • AUTH: STUDIO_REQUIRE_AUTH=true. The raw onrender URL requires the bearer STUDIO_AUTH_TOKEN (a Render secret; NOT the laptop .studio_token). Proven: /api/compose and /api/audio return 401 without it; /api/health, /api/pdf-check, /api/corpus-stats are public. The Cloudflare worker injects the matching token so studio.affirmology.ai still works (Access at the edge + token to origin).
  • PDF FIX: cloud PDFs were failing silently (pdf_url: None, audio still fine). Root cause via a /api/pdf-check self-test = ModuleNotFoundError: weasyprint - the agent pyproject never declared weasyprint (laptop-venv-only). Fix: added weasyprint>=60 to api/requirements.txt + native libs to the Dockerfile (libpangoft2-1.0-0, libfontconfig1, libglib2.0-0, libharfbuzz0b).
  • DEPENDENCY AUDIT (laptop venv vs declared deps, so no more silent omissions): also added pypdf>=4.0 - the post-render chart-integrity guard (scripts/chart_integrity.py) reads the PDF; once weasyprint produced PDFs the guard would have failed without pypdf. Everything else undeclared is corpus-build-only (bs4, trafilatura, readability, youtube_transcript_api, playwright, google-*, vertexai, tenacity, legacy PyPDF2) and intentionally NOT in the cloud image (render path proven not to need them). httpx is transitive via anthropic. elevenlabs is NOT imported at all (engine calls ElevenLabs over requests).
  • PROVEN RENDER (2026-06-18): Jeff / natal_sacred via the live cloud service with a valid token. Compose (council, corpus-grounded) ~137s/884 words; audio ~63s. Both artifacts confirmed in R2 by head_object: MP3 (10.2MB, audio/mpeg) + PDF (41KB, application/pdf). status: done with both URLs means the chart-integrity guard ran on the real PDF and passed.
  • DOMAIN REPOINT DONE (2026-06-18, Claude Code): the affirmology-studio-api Cloudflare worker now routes studio.affirmology.ai to the Render cloud. Set worker secrets STUDIO_FUNNEL_URL=https://affirmology-studio-api.onrender.com and STUDIO_AUTH_TOKEN (exact match to Render), deployed via wrangler (version c564c027). Verified worker->Render via the workers.dev URL before closing it: /api/health data_dir=/data, /api/people 200, /api/corpus-stats 12,905. Then set workers_dev = false + redeploy to CLOSE the side door (workers.dev now 404). The laptop server + Tailscale funnel were LEFT RUNNING as fallback (nothing retired); to roll back, set STUDIO_FUNNEL_URL back to the funnel URL and redeploy. studio.affirmology.ai/api/health returns 302 to Cloudflare Access (email OTP) so the final pass-through can only be confirmed in a browser (after Access login, data_dir=/data = cloud).
  • OPEN (Jeff/Cowork, needs R2/Cloudflare dashboard): R2 PUBLIC SERVING for media.affirmology.ai - rendered MP3/PDF are in the bucket but public fetch returns 403; wire R2 public access / the custom domain so the returned media URLs resolve. (The app domain repoint above is done; this is just the media host.)
  • ACCESS NOTE: Claude Code has NO Render/R2 dashboard access; it acted via git push (autoDeploy), R2 over boto3 with provided keys, and curl against the public endpoints. To have Claude Code do Render dashboard actions (restart, env, logs) provide a RENDER_API_KEY.

  20. Laptop housekeeping (2026-06-18, Claude Code) - Studio is on the cloud now

  Done now that studio.affirmology.ai runs on Render + R2 (section 19):

  • RETIRED the local laptop Studio server: stopped the FastAPI Studio process (uvicorn api.main:app on 127.0.0.1:8765). The Tailscale Funnel that exposed it was already off (Tailscale is STOPPED; funnel status = no config). Tailscale itself is LEFT INSTALLED, so re-exposing is a one-liner if ever needed. NOT TOUCHED: the nightly corpus crawl (there is no live launchd job anyway, ~/Library/LaunchAgents has none loaded; the plist is a Mac-mini-install template). The DEMO server (uvicorn server.api:app on :8443) was LEFT RUNNING (out of scope; it is the locked demo, not the studio).
  • MOVED THE CORPUS off the external SSD to the INTERNAL path /Users/jeffreyparker/CLAUDE/AFFIRMOLOGY/corpus (per Jeff: keep the research + db sorted under the AFFIRMOLOGY folder). Method: copy (ditto) -> verify -> remove. Verified byte-identical before deleting the SSD copy (same SHA-256 on corpus.db, same 6963 file count, PRAGMA integrity_check ok, 3866 docs / 12905 records). SSD copy then removed; studio-media on the SSD left intact. Contents moved: corpus.db, raw/ (581MB), extracted/ (57MB), logs/.
  • REPOINTED ALL PATHS to the new location (the literal /Volumes/Affirmology/corpus -> the internal path) across both engine trees (affirmology-agent and the vendored affirmology-studio/affirmology-agent): the engine default in script_generator.py, the crawl script scripts/overnight_corpus.sh, the nightly launchd/com.affirmology.nightly.plist (data-dir + log paths), and the corpus module docstrings. Also set AFFIRMOLOGY_CORPUS_DB=/Users/jeffreyparker/CLAUDE/AFFIRMOLOGY/corpus/corpus.db in affirmology-agent/.env. VERIFIED: the status module opens the new DB, and the engine's own db.search_records() returns rows from it (gene_keys/human_design/western all return hits). Module imports + py_compile clean.
  • HEADS-UP: the still-running DEMO server (PID on :8443) imported the OLD default path at launch, so until it is restarted a demo render would look at the now-removed SSD path and fall back to chart-only (robust, not a crash). The funnel is down anyway (Tailscale stopped), so no live demo renders are happening. Restart that server to pick up the internal corpus.
  • CORRECTED THE MISLEADING STATUS COUNTING. The raw "structured %" was distorted by NON-STRUCTURABLE RESIDUE the pipeline correctly skips: failed extractions (word_count 0/NULL: scanned books, papers, raw HTML) and sub-floor junk (<=100 words; the structurer's floor is >100). New rerunnable classifier affirmology-agent/scripts/classify_corpus.py writes the residue into a doc_quarantine table (idempotent; changes NO documents/records) and prints the corrected status; corpus/status.py now also prints a "Structurable status by tradition" block inline. Quarantined: 1,554 docs (1,474 failed extraction + 80 too short). CORRECTED structurable status (structured / structurable, where structurable = word_count>100 tier A/B):
  • human_design 967/1157 = 84% (residue 263 failed / 67 short)
  • transits 369/380 = 97% (0 / 2)
  • gene_keys 186/334 = 56% (77 / 2)
  • vedic_astrology 42/231 = 18% (290 / 6)
  • western_astrology 34/164 = 21% (383 / 3)
  • numerology 7/46 = 15% (461 / 0)
  • TOTAL 1,605/2,312 = 69% structured. The 707 pending is the GENUINE remaining backlog (vs the old ~12,905-record / 3,866-doc framing that made western/vedic/numerology look like 1-8% walls). Confirms PROJECT_STATE's long-standing note: the low percentages were never a budget wall, they were scanned-book/failed-extraction residue. LIMITATION: "raw HTML" residue that extracted with >100 words but is non-prose is NOT auto-detected (would need content inspection); flagged for a later pass. [UPDATE: see the recovery below; the 707 framing was itself too optimistic AND too pessimistic at once. Most "pending" was compressed-garbage that re-extraction recovered into real content, raising the genuine backlog to ~1,905.]

  Corpus corruption ROOT-CAUSED and RECOVERED (2026-06-18, Claude Code)

  A 10-doc Haiku pilot on the "pending" docs (to confirm they yield records before structuring all of them) returned 0 records. Investigation found TWO things: (1) the lowest-id pending docs are full scanned books and archive.org search/index pages (not interpretation prose), and more importantly (2) MOST "pending" extracted text was BINARY GARBAGE. ROOT CAUSE: the scraper saved the raw HTTP body still COMPRESSED (Brotli/gzip/deflate) without honoring Content-Encoding, and extract_html just .decode("utf-8")'d it, so the "extracted text" was garbage with a bogus word_count (passing the >100 floor). The raw bytes were intact, so it is fully recoverable with NO network calls. - FIX (permanent): added _maybe_decompress() to corpus/scraper.py (sniffs gzip/zlib/deflate/brotli, decompresses, leaves real HTML untouched); extract_html now decompresses before extracting. Mirrored into both engine trees, so future crawls are fixed too. (Brotli decompression needs the brotli package; the import is guarded so gzip/deflate still work without it. Declare brotli in the agent deps when convenient.) - DB PATH MIGRATION: the corpus move had left document.raw_path (3,865) and extracted_path (2,391) pointing at the deleted SSD prefix; remapped all to the internal path (0 left on SSD). - RECOVERY: new rerunnable scripts/recover_compressed.py (dry-run by default, --apply to write) decompresses + re-extracts every unstructured doc from its cached raw bytes and updates the extracted file + word_count when the result is real readable text. APPLIED: of 2,257 unstructured docs, 1,652 RECOVERED to readable text (numerology 492, vedic 384, western 372, human_design 263, gene_keys 141); 605 stayed unchanged (already-text or genuine index/placeholder pages); 0 raw files missing. DB backed up first to corpus/corpus.db.bak (reversible). - RESULT (classifier re-run after recovery): failed-extraction residue 1,474 -> 177; real structurable docs 2,312 -> 3,514 (1,609 structured + 1,905 pending); total quarantined residue 352. The corpus is far healthier than the old framing assumed. - CORRECTION to the prior-session claim that "the remaining backlog is non-structurable junk": that was largely WRONG. The bulk was compressed-but-recoverable real content; ~1,200 docs were salvaged. A small genuine residue remains (index/search/placeholder pages + a few un-decompressable files). - PILOT PROOF that recovered docs structure: re-structuring a sample of recovered pages with Haiku yielded real records (Chaldean numerology 14, Cafe Astrology 11, gene keys 7, etc.). Total pilot API spend ~$0.67 (Haiku). - NEXT (needs Jeff's go; this is the credit-spending step): run the full Haiku structure-only pass over the ~1,905 pending docs now that they are real text. Rough estimate ~$15-40 at Haiku rates depending on book chunking. Recommend tradition batches (numerology/western/vedic first, where recovery added the most) with --max-cost-usd caps and a small --limit re-pilot per tradition to confirm yield before each full batch.

  CORPUS DIRECTION CHANGE (2026-06-18, Jeff + Cowork) - ACQUISITION over re-grinding

  Standing plan: Affirmology_CorpusAcquisition_Plan_v1.md (hand to Claude Code, runs unattended within budget). Jeff's call after reviewing Code's Western run: the corpus feels weak because effort went into re-structuring a scrape full of archive.org search?query= / index / junk pages (the "6 Western big books" were all search-result pages, 0 yield). The leverage is ACQUIRING real, legal full text and parsing THAT. - PRIORITIES: Western (highest) > Gene Keys + Vedic (both important) > numerology SHELVED (lightly used; do not spend on it now). - THE UNLOCK: fetch public-domain (Tier A) full-text books, not search pages. Western: Gutenberg (Ptolemy Tetrabiblos #70850, Sepharial #46963) + archive.org *_djvu.txt (Raphael, Alan Leo, William Lilly). Vedic: sacred-texts.com Jyotish + PD Parashara. Gene Keys: A/B only, TIER WALL on Rudd/Ra Uru Hu (Tier C, never scraped); use PD I Ching (Legge, not 1950 Wilhelm) + A/B community commentary. - MODEL/COST: Gemini key now works, so re-open the structurer choice (old "HOLD on Haiku" was the Gemini billing block). Run a 10-doc bake-off Gemini Flash vs Haiku vs Nemotron; use the cheapest good model for the bulk. Jeff has Gemini + Nemotron + Haiku credit. - BUDGET: Jeff adding ~$15-20 now, re-ups when proven useful; pilot-before-batch, hard --max-cost-usd caps, quarantine noise permanently, stop at ~$3 floor and report. - PUSH after every significant piece (each tradition/big-book set overwrites R2); ONE Render restart at the end, not after every push. - COORDINATION: Code stays in affirmology-agent corpus code; do NOT touch affirmology-studio or numerology.py/council.py/script_generator.py (Cowork editing in parallel). The engine items (numerology guard, Persephone gate) land AFTER the corpus run is done and pushed.

  21. Corpus ACQUISITION run (2026-06-18, Claude Code) - real public-domain books + Gemini Flash

  Per Affirmology_CorpusAcquisition_Plan_v1.md. Stopped re-grinding the old scrape; ACQUIRED real full text and parsed it. New corpus scripts (all in affirmology-agent/scripts/, corpus code only): acquire_books.py (Gutenberg + archive.org _djvu.txt, never /search), bakeoff.py, gemini_structure.py (concurrent, tag-validating), run_plan.py, coverage_report.py, classify_corpus.py, recover_compressed.py, structure_batch.py, pilot_v2_prompt.py. - MODEL BAKE-OFF: Gemini Flash (gemini-2.5-flash) WON decisively: 33 records / 2 book chunks for $0.0011 vs Haiku 3 records / $0.042. Nemotron yielded 0. Gemini is the bulk model now (key works; old "HOLD on Haiku" is lifted). ~400x cheaper/record. Default Gemini, escalate only if a pilot underperforms. - WESTERN: acquired 6 PD books (Ptolemy, Sepharial x2, Raphael, Alan Leo, William Lilly). KEY FINDING: record quality == SOURCE quality, because the structurer prompt preserves the source's voice (rule 2). Pre-1900 OCR books (Lilly 1647, Raphael 1828, Ptolemy) produce archaic + OCR-garbled records -> QUARANTINED (not deleted) into the new record_quarantine table (2583 + pilots = 3682 rows), their sources marked DEFERRED, docs reprocessable. Modern-era books (Alan Leo ~1910, Sepharial ~1920) produce clean evergreen records -> KEPT. Western book records added this run: Alan Leo 540, Sepharial Manual 479, Sepharial How-To 427 = ~1,446. Western total 4,297 -> 5,743 (depth 11.4). - GENE KEYS + VEDIC: book acquisition hit UNRELIABLE archive scans - the "Legge Yi King" DLI scan (in.gov.ignca.20915) is mislabeled (Western astrology body), and "Brihat Jataka 1885" (wg1079) is a bound miscellany mixing Brihat Jataka with Western directions. All pre-1900 anyway. DEFERRED to v2 + source verification. Gene Keys stays 670, Vedic 391. Tier wall HELD: never scraped Rudd/Ra Uru Hu (Tier C); only public-domain texts attempted. - COST: the whole acquisition run cost ~$0.30 (Gemini Flash). The fresh $20 is essentially untouched. Budget is no longer the constraint; source quality + correct sourcing are. - CORPUS TOTAL: ~20,060 records. corpus.db pushed to R2 (15.8MB, 2026-06-18 17:12 UTC). NEEDS ONE Render restart of affirmology-studio-api to go live (re-pulls on boot when R2 size differs). - v2 PROMPT DRAFTED + PILOTED (NOT yet used for reprocessing, awaiting Jeff's go): pilot_v2_prompt.py holds a v2 structurer prompt that EXTRACTS THE MEANING in clean modern neutral English instead of preserving archaic wording (unlocks the deep classics; safer on copyright). Piloted on William Lilly 1647: faithfully modernized garbled 1647 OCR into clean records (e.g. source "if the major part of the Significators be in Diurnal Signs, the Birth will be by day" -> "If the majority of chart significators are placed in diurnal signs, the birth will occur during the day"). Faithful, not fabricated. NEXT (on Jeff's go): adopt v2 and reprocess the quarantined classics (Lilly/Raphael/Ptolemy) + re-source verified I Ching (Legge) and Brihat Jataka for gene_keys/vedic.

  v2 prompt ADOPTED + classics reprocessed + Gene Keys/Vedic re-sourced (2026-06-18, later)

  • v2 STRUCTURER PROMPT adopted in structurer.py (extracts MEANING in clean modern neutral English, not source's archaic voice) + brand-safety rule (omit derogatory race/caste/gender wording, skip mundane nation astrology, keep individual chart interpretation). Piloted faithful (not fabricated) across Lilly/Raphael/Ptolemy before reprocessing.
  • WESTERN classics reprocessed on v2/Gemini: Lilly 2,149 + Raphael 370 + Ptolemy ~0 (its text is mostly mundane/ethnographic, correctly filtered). Western total 4,297 -> 8,219 records (depth 16.3). The archaic v1 records stay in record_quarantine.
  • GEMINI CREDITS DEPLETED mid-run (separate Google AI Studio billing; NOT the $20 Anthropic top-up; total Gemini spend this session < $1). Escalated Gene Keys + Vedic to Haiku per the plan's "escalate when cheap model underperforms."
  • SOURCE BUG FIXED: dense cookbook chunks (Brihat Jataka) overflowed Haiku's max_tokens -> truncated JSON -> 0 records. Fixed: max_tokens 4096 -> 8192 + truncation-salvage in _parse_and_persist (keeps complete objects from a cut-off array).
  • GENE KEYS re-sourced to VERIFIED genuine Legge Yi King (archive wg916, SBE Vol 16 1882; content-checked 1407 I Ching hits / 0 western). +1,009 records -> 1,679 (depth 20). Tier wall held (Legge PD; never Rudd/Ra Uru Hu).
  • VEDIC re-sourced to VERIFIED genuine Iyer Brihat Jataka (archive brihatjatakavar00iyergoog, doc 3875; 717 vedic hits). The earlier wg1079 was a bound miscellany (mixed Western) and in.gov.ignca.20915 ("Yi King") was a mislabeled Western scan - both DEFERRED. +1,335 records -> 1,726 (depth 38.4).
  • RECORDS ADDED THIS ACQUISITION+REPROCESS PASS: western ~+3,922, gene_keys +1,009, vedic +1,335. CORPUS TOTAL ~24,949 records (was ~12,905 at the cloud baseline). Anthropic spend for gene_keys+vedic Haiku ~$1.69; fresh $20 barely touched.
  • corpus.db pushed to R2 (18.2MB, 2026-06-18 18:02 UTC). NEEDS ONE Render restart of affirmology-studio-api to go live.

  Polish run: numerology + Part of Fortune + asteroids/points + Gene Keys layer-b (2026-06-18, later)

  Proven cadence (verify source genuine + public-domain -> v2 modernize/de-bias -> pilot shown -> scale). Gemini credits depleted (separate Google billing), so all on Haiku. - NUMEROLOGY (was shelved, un-shelved with verified PD classics): Westcott "Numbers: Their Occult Power & Mystic Virtues" (1890) + Sepharial "Kabala of Numbers" (1920), content-verified. Cheiro skipped (no clean PD scan). +514 records -> 1,138 (depth 18.7). Yields number symbolism (Number_1..9 + masters), the basis for life-path/expression retrieval. - PART OF FORTUNE: already well-covered (~19 modern records, houses 1-12 + aspects); classical Ptolemy/Lilly extraction mostly deduped against the better modern set. Done. - ASTEROIDS + POINTS (Chiron, Ceres, Pallas, Juno, Vesta, Black Moon Lilith): modern, not in PD classics, so synthesized from MULTI-SOURCE A/B web consensus in our OWN voice (no single-site verbatim), general + 12 signs + 12 houses per body. +51 new (the rest of the keys already existed). scripts/synth_points.py. - GENE KEYS LAYER-B (navigation methodology, tier-walled): scripts/synth_genekeys_b.py. 19 methodology/navigation concept records (sequences/spheres/contemplation/three-frequency model, our words, NOT Rudd's Activation/Venus/Pearl names) + shadow/gift/siddhi for 61/64 keys, each DERIVED FROM THE PUBLIC-DOMAIN I CHING (King Wen) hexagram theme in our own words. TIER-WALL FIX: quarantined 180 pre-existing shadow/gift/siddhi records that came from a third-party blog reproducing RUDD'S NAMED SET verbatim (e.g. "Gift of Guidance" GK7, "Shadow of Inertia" GK9) into record_quarantine (reason genekeys_rudd_named_tierwall); replaced with our clean I-Ching-derived set (verified: GK7 gift = "strategic coordination" not "Guidance"). Kept shadow/gift/siddhi only as generic frequency labels per Jeff. - FINAL CORPUS (2026-06-18): western 8,271 / human_design 7,952 / transits 4,764 / vedic 1,726 / gene_keys 1,701 / numerology 1,138 = ~25,552 records. corpus.db pushed to R2 (18.6MB, 2026-06-18). NEEDS ONE Render restart of affirmology-studio-api (Claude has NO Render API key; Jeff restarts manually) to go live; service re-pulls corpus on boot when R2 size differs.

  17. Wealth codes reading + paste-ready blueprint data + stored Vedic (2026-06-17)

  • CLAUDE OUTPUTS/Affirmology/Affirmology_WealthCodes_JeffAndSol_v1.md: individual prosperity readings (Gene Keys Pearl WITH line data, HD wealth gates/channels, astrology money houses, full-name numerology), the block/shadow, near + long-term actions, 2026 timing, energetic blueprint for resources/abundance/impact, plus a joint money-machine layer.
  • CLAUDE OUTPUTS/Affirmology/Affirmology_BlueprintData_JeffAndSol_PASTE_v1.md: ALL chart data for both, plain/labeled, copy-paste-ready for Sol to drop into AI tools (incl. 2026 transits).
  • Vedic STORED as engine facts: affirmology-agent/out/jeff_speed_08/vedic_facts.md and out/soledad_gabriela_ballard_1986-01-23/vedic_facts.md (sidereal Lahiri: signs, degrees, nakshatra+pada+lord, lagna). Delivery copies: Affirmology_VedicProfile_{Jeff,Sol,JeffAndSol}_v1.md. Both Moons in Ardra (Gemini, Rahu); Sol heavily Capricorn sidereal; both Mars Scorpio 0 deg. Numerology confirmed from full names: Jeff Expression 1 / LP22; Sol Expression 3 / LP3 (triple-3).

  18. First full agentic wealth readings via /api/hermes (2026-06-20)

  The CLAUDE.md rule is now LIVE in practice: astrology routed through the agents, not the chat. Ran five oracle readings over the funnel and compiled them: CLAUDE OUTPUTS/Affirmology/Affirmology_WealthCodes_OracleReadings_v1.md (~43k chars): Jeff x Lakshmi, Jeff x Agastya, Sol x Lakshmi, Sol x Agastya, Together x Concordia. Quality is a clear step up: mechanism-level, multi-system synthesis, uses the new points (Black Moon Lilith, Chiron, Part of Fortune), PROGRESSIONS (e.g. Sol's Progressed Moon in Capricorn), Vimshottari dasha, and 2026 transits. Jeff = "Merchant-King" (8th-house capital + Gate 45); Sol = "identity not hustle" (NN + Lilith Taurus, 11th-house community); Agastya read past money-hardship via the Ardra/Rahu Moon and dasha. OPERATIONAL NOTES for the agents: (a) single/1-2-oracle calls work reliably; MULTI-oracle (6+) still hits the residual forced-synthesis BadRequestError (see Affirmology_HermesLoop_Bug_v1.md) - so run focused calls and compile. (b) /api/hermes is synchronous and slow (>45s); POST then poll GET /api/hermes/history (the reply persists even if the client times out). (c) heavy threads get flaky as history grows; a couple calls hung/stubbed before succeeding on retry. (d) minor accuracy wobble seen (Jeff reading briefly conflated Pearl 44.1 with Radiance 45.5) - exactly what the planned craft-critique QA pass should catch.

  19. Before-bed WEALTH audios on the new Quartz Pulse Tone bed (2026-06-20)

  New sleep bed added: affirmology-agent/music/Quartz Pulse Tone.mp3 (Jeff uploaded; ~8.6 min, 192k). Use MP3 for beds (pipeline re-encodes to mp3; WAV master kept by Jeff). Category for the bed selector: Stillness/Delta (before-sleep). Rendered 2 before-bed wealth/abundance audios (theme: call in abundance, get Affirmology funded, comfort in path + timing). Method: compose before_bed via Studio on the Mac (chart-driven), grab the engine's voice-only intermediate (backend/*_voice.mp3), re-mix with Quartz in the sandbox (voice_offset 8, music 0.18, 9s fade). This is how to use a NEW bed before the Studio bed-selector ships. Delivered to CLAUDE OUTPUTS/Affirmology/KickoffWeek_Audio/: Jeff_BeforeBed_Wealth_Quartz_v1 (9:07), Sol_BeforeBed_Wealth_Quartz_v1 (7:02).

  20. Investor-video TITLE SYSTEM + name-morph (2026-06-24)

  Built the reusable gold AFFIRMOLOGY title system for the movie-trailer bridge. Winning pipeline (locked): generate the gold title as a STILL (Nano Banana, perfect text) → animate with Higgsfield Kling 3.0 two-keyframe morphing (start_image + end_image) → key out the black (lumakey) for transparent overlays. Hand-coded/remotion text was abandoned (looked cheap). Full spec: CLAUDE OUTPUTS/Affirmology/Affirmology_TitleSystem_v1.md. - CANONICAL frame (the alignment lock): AFFIRMOLOGY_canonical_BLACK.png + _ALPHA.png (1280x720). Every title animation starts/ends on this so they line up pixel-perfect when stacked in CapCut. - Assets in Affirmology_VideoEditKit/grand_bridge/TITLE_SYSTEM/: INTRO_black_to_AFFIRMOLOGY (.mp4/_ALPHA.mov), OUTRO_AFFIRMOLOGY_to_black (.mp4/_ALPHA.mov), affirmology_NAMEMORPH_v4_snappy (.mp4/_ALPHA.mov = black→AFFIRMOLOGY→Affirmations+Astrology+AI→Affirm+ology+AI→Affirmology.ai→AFFIRMOLOGY). - Also in grand_bridge/: nebula entrance pieces (nebula_bloom, title reveal, lockup_living), soul-song + subconscious-OS transparent overlays, hero constellation light-up v3 (the painterly hero figure igniting). All on-disk for Jeff to assemble in CapCut. - Jeff drives final assembly in CapCut (the agreed workaround); chat generates premium assets + transparent overlays. Next options: more intros/outros anchored to canonical; rebuild nebula entrance off canonical for full alignment; push morph even faster.

  21. Master edit guide + oracle browser prompt pack (2026-06-24)

  • MASTER EDIT GUIDE for the Joe & Monika investor video: CLAUDE OUTPUTS/Affirmology/Affirmology_InvestorVideo_MASTER_EditGuide_v1.md. The single map: story-spine clips 01-24, new scenes (rooftop guide, dog-walk, Monika arc), the title bridge (TITLE_SYSTEM is ONLY the gold-title bridge, not the whole film), all overlays, and the audio files (master_voice_track / VOICE_ONLY / music_bed in VideoEditKit/audio). Flags gaps: new VO (guide dialogue, Monika Sacred Audio, Atlas thesis), the 2 Seattle zero-proof company clips.
  • NAME-MORPH set expanded: chrome v4 gold + ink (keepers), plus build-on-from-parts versions with STYLE-MATCHED intros (ink blooms in, shards assemble in): affirmology_NAMEMORPH_INK_buildON_v2, affirmology_NAMEMORPH_SHARDS_buildON_v2 in TITLE_SYSTEM. All transparent _ALPHA.mov, anchored to canonical frame.
  • ORACLE CHARACTER SHEETS - free browser path: CLAUDE OUTPUTS/Affirmology/Affirmology_OracleSheets_BrowserPromptPack_v1.md. 14 remaining oracles (Agastya, Prometheus, Athena, Pythagoras, Orpheus, Apollo, Hypnos, Chiron, Persephone, Echo, Eros, Concordia, Hestia, Gaia) as paste-and-go hero-bust prompts in the locked marble-and-gold Sophia look, runnable FREE in browser Nano Banana 2 (attach locked Sophia frame as reference). Run options: Jeff pastes, or chat drives Chrome. Then full turnaround sheets + Higgsfield Character Elements.

  22. Investor-video VOICE CAST (locked record) + new VO rendered (2026-06-24)

  FOUND in Higgsfield history (so we never lose it again). The v7 character VO uses the inworld_text_to_speech model with these voices: - JOE = Ethan (en) - MONIKA = Olivia (en) - FRIEND/guide = Liam (en) - In-app ORACLE = Serena (en) - ATLAS (bridge thesis) = Arthur (ElevenLabs preset, voice_id 30fc8796-ceb6-4a66-b3a7-4a145ef7f346) - NOT inworld; locked this session. - Jeff's personal wealth audios (separate) = Orion (ElevenLabs, ed69c516). Rule note (Jeff 2026-06-24): the "all audio through our engine" hard rule is for PRODUCT Sacred Audio; the INVESTOR VIDEO VO is allowed via Higgsfield to match prior choices. NEW VO rendered + saved to Affirmology_VideoEditKit/audio/new_vo/: joe_guide_kickoff (Ethan); friend1_partytrick / friend2_testimony_seattle / friend3_reframe (Liam); oracle_wealthblueprint (Serena); monika_SA1_hours / SA2_circle / SA3_ceiling / SA4_worth / SA5_becoming (Olivia); atlas_thesis (Arthur). m_ceiling stays as-is (already Olivia). Personalization through-line: Affirmology_PersonalizationTagging_Review_v1.md + 6 new source-tag overlays in VideoEditKit/overlays/ (tag_joe_ and tag_stars_). STILL OPEN: the 2 Seattle zero-proof company clips (factory cheers + laptop success) - need the blond-friend anchor image; prompts in Affirmology_NanoBanana_PromptPack_v1.

  23. The FEMALE TAO - locked keeper image (Jeff, 2026-06-24)

  The female Tao (Wisdom Circle; the receptive yin, water, yin-yang, jade; absorbs Quan Yin). - LOCKED KEEPER = the FORWARD / screen-facing take: hf_20260618_223234_8ca0b8be-e199-4998-afad-5cd02351b31f.png → https://d8j0ntlcm91z4.cloudfront.net/user_3F3pFGHoR8fJGcZbldgyQ5ENom4/hf_20260618_223234_8ca0b8be-e199-4998-afad-5cd02351b31f.png - Morph/avatar alt (parked idea): the more angled three-quarter take hf_20260618_223234_c6a448b9-248d-43d2-b81f-b9317af07175.png - Jeff: she could MORPH into this as an avatar variant. Not the primary. - CORRECTION: older sheets (Affirmology_WisdomCircle_Explorations_v1.html, Affirmology_VisualBible_v1.html) had mislabeled c6a448b9 as "screen-facing LOCKED"; the actual forward-facing keeper is 8ca0b8be. Records flipped to match.

  ---

  2026-06-27 - PDF Report v2 direction locked (Cowork session)

  The demo PDF is being redesigned from a 5-page audio liner note into the logical-brain asset of the demo trio (Soul Song audio = emotional, reading audio = trust/learning, PDF = logical proof they keep and reopen for months, and a marketing asset). Full flow in Affirmology_PDFReportV2_Architecture_v1.md; implementer build spec in Affirmology_PDFReportV2_BuildSpec_v1.md; reading craft in Affirmology_AhHaMoves_Registry_v1.md. Locked decisions: next-morning delivery (with the reading audio); content/prose first then a light visual pass with hero/closing image placeholders (Jeff creating those); reading generated through the council pipeline with a second-agent recognition cross-check; progressions excluded (in-season layer reserved for the paid app); our own branded SVG bodygraph/wheel/Gene-Keys/Vedic/numerology on a single "Blueprint at a Glance" page; multiple QC gates with auto-redo and a human-review hold so nothing broken ships (PDF analog of scripts/audio_qc.py); run everything through the real pipeline, not local renders. Test cohort: Jeff, Colin, Josh Parini. Working cover title: "Your Origin Report." Separate workstream noted: tokenized permanent landing page per person (Soul Song player + reading + marketing + permanent PDF link).

  2026-06-27 (later) - Origin Blueprint PDF v2.1 revision + marketing profile

  v2 cohort PDFs (Jeff, Sol, Josh, Colin) rendered and passed QC; prose is gold, but Jeff's review: too brief on every section and under-designed (half-beige, blank half-pages). Full revision in Affirmology_PDFReportV2.1_RevisionBrief_v1.md. Title changed to "Your Origin Blueprint." Mandates: deeper/explanatory/persuasive sections (stop optimizing for brevity), correct persuasion psychology (no presumptuous "you were never broken/lost"; use "you may have"; explain jargon before using it; set the why first like the investor video), commit fully to the locked emerald-and-gold brand system with embedded Cormorant/Inter/JetBrains fonts, real logo (Affirmology_Logo_SealMark_v1.svg) and real Oracle art (affirmology-app/shared/assets/oracles/*.png, Council/OracleSheets/*.png) instead of placeholders, the quadrant Blueprint page (Western/HD/GeneKeys/Vedic quadrants + numerology diamond, each with our own branded chart graphic + Oracle tag), a rebuilt Method page selling from our research (predictive brain, repetition, theta, heart coherence, market stats) and per-system primers, a rebuilt Path Forward (Identify/Align/Transcend, app pitch with audio contexts, membership categories Relationships + Wealth Codes, Oracle feature mention, contextualized teaser cards), 18-22 full pages, no blank pages. New QC gates: anti-brevity, anti-presumption, explainer, design/density. Code NOT committed yet (still active redesign). NEW WORKSTREAM (capture now, use progressively): per-person marketing profile from chart (Sun sign for ad targeting, HD type/authority for nudge cadence and CTA, loudest themes for which teaser to lead) plus in-app engagement signals, stored on the lead record with email/phone/token, powering week-one nudges and ad retargeting.

  ---

  LOG 2026-06-28 - Investor White Paper full first draft

  Affirmology_InvestorWhitePaper_v1.html now a complete first-pass: cover (locked two-orb PNG, soul + system headlines, "Affirmations + Astrology + AI"), Executive Summary, and 13 Parts (Vision, Problem, Science, Product/System, Market, Go-to-Market/Loft/creator, Rollout, Team, Astrology of the Moment, State of Project, Moat, The Ask + economics table, Future). Applied this session: cover reverted to short "Upgrade Your Subconscious Operating System"; user app renamed "the Affirmology user app" (Maya dropped); synthesis-and-poetry presentation layer named; July 4 beta enrollment made explicit; cap table = Jeff >50% control (holds CEO+employee reserves), Sol 25% vesting, 5-person board (Jeff/Sol/Colin+2), full cap table on request, vesting not published. Zero em dashes. PENDING: Jeff to riff his own voice into each section; insert Colin bio; add image plates (scene stills + event photos + character art); incoming transparent-bg logos + font/brand doc to apply; get a draft to Colin.

  LOG 2026-06-28 (b) - White paper expanded to 16 parts + stats verified

  Added 3 product-depth sections: V The Constellation Effect (gamification/retention done honestly), VI Data and Intelligence (living profile + flywheel + future voice), VII The Journeys (upsell ladder + creator program). Renumbered to 16 parts; Future is now Part XVI. Embedded 5 real photos (2 founders, astrology hotline, HD-workshop teaching, Aurea crowd). Cinzel wordmark on cover. Council line corrected to Sri Agastya (Vedic), Parashara removed. Pre-sleep reframed as best-but-not-only window. STATS: ran web verification. DROPPED as unsubstantiated: $30B ocean, 48M affirmation listeners, 83% don't-stick, 3M HD/GK practitioners. CORRECTED: Calm ~$227M, Headspace ~$348M (2024), Co-Star 30M+ registered, astrology affinity 62-83% (Harris 2024). Full reconciliation appended to Affirmology_WhitePaper_FactsPack_v1.md. Graphics briefs flagged to use only verified figures. Still pending: Science-section images (clockwork-head + strengthening-pathway) to be dropped into Brand/photos/; Colin bio; new brand assets/transparent logos to apply.

  24. CHARACTERS master index (2026-06-24)

  One consolidation for all character work (council oracles + Wisdom Circle + hero sheets + Olympus environments): CLAUDE OUTPUTS/Affirmology/Affirmology_CHARACTERS_MasterIndex_v1.md. Points to the 3 handoff docs (OracleHeadPhotos, WisdomCircle, OracleFullSheets cross-chat brief), the master visual records (VisualBible = source of truth for which image is whom, FullCast, VisualGallery, Lookbook), the roster/style docs, and the per-character sheets. Includes inline locked links: Atlas + Sophia model sheets, the Tao keeper, the 4 Olympus environments. Share this file with the app/other chats.

  ---

  2026-07-06 - Business Plan v3 (founder voicenote answers assimilated)

  Jeff voicenoted answers to all 14 GapWorkbook sections; folded into Affirmology_BusinessPlan_v3.docx + Affirmology_FinancialModel_v2.xlsx (v1 was locked/open, so model saved as v2; plan references v2). Google Doc: https://docs.google.com/document/d/1BRB1IQvxTcHmw8oyJ1z_qGSv5iGUXXRh6GaS6eSgEbQ/edit Key decisions locked this round: - Cap table: Jeff ~75% (old 8% pool folded conceptually into Jeff's stake; grants come from Jeff or mutual-consent dilution). Sol 25% target = 10% vested + 5%/yr, purely TIME-BASED (not investor-tied). Colin path to ~5% vested w/ defined terms. Randy: per-event pay, NO equity promised (pool candidate). Jacque CEO ~10%+ later, triggered ~$100K/mo revenue, $20-25K/mo. - Governance: NARROWED Sol's unilateral domain to the social accounts she runs (IG/TikTok); YouTube shared; brand/visual collaborative, set in operating agreement not declared. REMOVED the 50/50 reinvest/distribute rule (keep only 6-mo reserve then quarterly at board discretion). - Distributions/round: goal is NO Round 2; not selling >20% to investors; likely close well under $150K. - Founder pay $8K/mo combined now, rising; less drawn at low raises. - Faena: VERBAL only, signing this week (not "confirmed"); daughter is on-site assistant; goal 20-200 signups (up to ~400), ~30 affiliates + VIP first-night. - Beta opens JULY 10 (not July 4). Targets: >=500 subs by Nov, >=2,000 by end Q1'27. - Voice cost honesty: ElevenLabs ~$0.50/audio now; grant zeroes yr1 (submitting this week via demo URL); long-term grow on Fish Audio (cheaper, also grant). Music: only commercially-cleared beds (Suno/purchased) + seeking custom-bed partnerships. - Wynwood loft trigger ~$20-25K/mo revenue. International: Spanish first, within yr1. Provisional patent: Jeff (patent attorney) files, gifts to company. - Creator mechanics still TESTING (rev-share + credits/perks + explicit affiliate deals for big accounts). Warm creator relationships to activate later: Danielle Page, Caitlin (names kept OUT of investor doc). Event staff: Lauren Martinez (early mgr ~$1K/mo + event pay), Glo Ruiz (clerical), $150-200/day event rates. - Use of funds added: pro headphones (~200 @ ~$16 + splitters), audio beds ($100-200 ea, ongoing), storage unit + transport carts, staff shirts, capture gear + Sol's home content set + video-editing hire, Art Basel (optional Dec), Claude Team ~$500/mo. - Investor perks beyond financials: free event access, promote their center/community; $5K+ tier gets more voice/favorable treatment. STILL OPEN: beta metrics (post 7/10), final tier prices + creator rev-share, countersigned Faena, trademark filing, operating agreement w/ counsel, Round-2 terms if any. Term sheet with $75K tier still to finalize (awaiting Jeff's extra notes).

  ---

  2026-07-08 - Plan v4 (readable rebuild + growth engine), Norm Adams ask, Sol social + affiliate docs

  • BusinessPlan_v4.docx: fully reformatted for readability (branded title block, Georgia headings w/ gold rules, bold lead-ins, clean shaded tables, spacing). Jeff said v3 formatting was "wildly unreadable" - fixed. Rendered to PDF and visually verified pp.1-2.
  • Added GROWTH ENGINE section (Sec 8) + TWO STAGES OF CAPITAL (Sec 12) + growth-economics note (Sec 14), grounded in research: Co-Star ~20M+ downloads on ~no paid (shareable push notifications), CHANI ~$14M rev / no VC (founder+email+values), payback ~2 mo at $27/$50, reinvest 20-40% of rev (Astrotalk ~25%), churn is the master lever. Two-stage raise: Stage 1 prove engine, Stage 2 growth capital as ad fuel (RBF 1.2-1.6x cap / venture debt / growth equity), lower risk, better terms, early investors get preferred position to add fuel.
  • Model v2: added 'Growth & Financing' tab (LTV/payback by churn, reinvestment flywheel table, Stage-2 financing options). Recalc 0 errors.
  • Plan v4 Google Doc: https://docs.google.com/document/d/1bgpXqIHauPFYXNkCV7NJmr12bSxUaXuQvD6G-1ED69E/edit
  • NORM ADAMS (same-birthday, runs retreat center "Coherence"): interested, may do 25/30/50 + connect future investors for ad-scaling. His concern: sustainability (don't waste his $ if rest isn't funded). Built Affirmology_NormAdams_UseOfFunds_v1.docx - a 1-page $30K focused ask: Founder focus Aug+Sept $16K, app help $3K, Sept events $5K, ad testing $3K, tech+music beds $2K, buffer $1K = $30K; framed as self-contained bridge to launch+first revenue; Stage-2 ad money is separate/lower-risk (where Norm can help). Google Doc: https://docs.google.com/document/d/1cY9g-6gBztHrWHNkbFHDI9BiTFlabpesxDeqd9bo8Hw/edit [SEPARATE ongoing mini-project per Jeff; will iterate.]
  • Sol social playbook (Affirmology_SocialGrowth_Playbook_ForSol_v1.docx): Co-Star + CHANI organic/creator growth teardown + 12 copyable tactics + IG/TikTok 2025-26 + 30-day starter plan.
  • Affiliate rewards model (Affirmology_AffiliateRewards_Model_v1.docx): credit-first 3-tier - Members (audio credits + comped months), Creator Accounts (status/access/comped premium), Paid Influencers (25-35% recurring rev-share funded from referred revenue). Matches Jeff's instinct exactly. STILL OPEN: refresh Sol/affiliate as Google Docs if wanted; term sheet extra notes; beta metrics post 7/10.

  ---

  2026-07-12 - Demo Experience doc + DEFAULT DOC STYLE LOCKED

  • Affirmology_DemoExperience_v4.docx is the final demo-area doc for Norm. Opens with Jeff's "Imagine this" narrative under the hero render; chaise-with-mat-built-in gets its own section; adds "Where this goes next" (foot baths, essential oils/diffusers, scent at song start; chant/bowls explored not promised) and "Someday: the Affirmology Loft" (biohacking, focus, sound+light, breath/coherence, recovery + Zero-G chair, Vibroacoustix). Products live-linked in prose. Unknown costs = "coming soon". Headphones $12 ea in bulk. Rugs/bean bags: borrowing Giselle's to start.
  • Verified prices: Woojer Vest 4 $479, Woojer MAT $999, Lumenate Nova ~$105 (£82.50), Eames $459, chaise $375.
  • Concept renders (original art, NOT vendor photos, which are copyrighted) in Demo/renders/. Hero = emerald Eames + the approved chaise + bean bags + starry projection ceiling at an event.
  • STYLE LOCKED: see CLAUDE.md "DEFAULT DOCUMENT STYLE" + Ops/affirmology_docstyle.py. Use it for every doc.
  • KNOWN GOOGLE DOCS BUGS (root cause of the earlier "unreadable" docs): Docs drops external , and turns bold/links inside table cells into literal \*\* and \[text\](url). Never do either.

  ---

  2026-07-12 - FEDERAL EIN ASSIGNED

  EIN 42-3770990 assigned to AFFIRMOLOGY LLC (name control AFFI), classified by the IRS as a MULTI-MEMBER LLC (default partnership tax treatment, Form 1065 + K-1s; accounting year closes December). Formation: Wyoming, Laramie County, effective June 26, 2026. Physical: 1621 Central Ave, Cheyenne WY 82001. Mailing: 1444 NW 14th Ave Apt 1809, Miami FL 33125. Full details in Affirmology_CompanyRecord_v1.md (CONFIDENTIAL - never put the EIN in investor decks or shared docs). UNBLOCKS: business bank account (bring EIN letter + Articles + Operating Agreement), Stripe/merchant processing. STILL OPEN: save the IRS CP 575 confirmation PDF into this folder; finalize the Operating Agreement with counsel.

  ---

  2026-07-12 - OPERATING AGREEMENT DRAFT + BANKING

  • Affirmology_OperatingAgreement_DRAFT_v1.docx drafted. SUPERSEDES Affirmology_CoFounderAgreement_v1.html (May framework). Key supersessions: Jeff ~75% (pool is a concept from his units, NOT a reserved 8% block); Sol 25% target = 10% VESTED AT SIGNING + 5%/yr for 3 yrs, PURELY TIME-BASED (old investor-bring-in earn-in is DEAD); Sol's domain NARROWED to the social accounts she runs (YouTube shared, brand collaborative); distributions = 6-mo reserve then quarterly at Board discretion (NO fixed 50/50); Colin path to ~5% w/ requirements TBD; Randy no equity; Jacque CEO ~10%+ at ~$100K/mo revenue. KEPT from old: Major Decisions list, romantic dissolution clause, good/bad leaver + 50% acceleration, mediation ladder, IP assignment, confidentiality/non-disparage/non-compete/non-solicit, ROFR/tag/drag, $5K/$25K finance tiers. ADDED NEW: mandatory TAX DISTRIBUTIONS (partnership taxation means phantom income risk), banking authority section.
  • Affirmology_BankingResolution_v1.docx - one-pager for the bank's file.
  • LEGAL FACTS VERIFIED: WY does NOT require an OA and it is NOT filed with the state (internal doc). Notarization NOT legally required, but some banks ask for it (recommend notarizing anyway). Domestic LLCs are currently EXEMPT from FinCEN BOI reporting under the Mar 21 2025 interim final rule (still in effect 2026).
  • OPEN BEFORE SIGNING: member- vs manager-managed (must match Articles); Sol's vesting start date; Colin's exact terms; 3rd board seat; Sol's dept budget; tax-distribution rate (drafted 40%); governing law (WY vs FL) + arbitration venue; counsel review of non-compete.

  ---

  2026-07-12 - OA v3, ENTITY ARCHITECTURE, BUSINESS PLAN v6

  FOUNDER EQUITY (Jeff's call, supersedes all prior): - Sol: 10% UNCONDITIONAL at signing (the past/partnership/blessing; never clawed back, survives romantic split). +5% at yr1, +5% at yr2 = 20% on TIME ALONE. Then a GROWTH LADDER to 25%: 1% each at 10K / 25K / 50K / 100K / 250K combined ORGANIC followers on the COMPANY-owned IG + TikTok (social@affirmology.ai). No expiry, no clawback. FRAMING (important): present to Sol as "you have 20%, plus a bonus ladder to 25%", NEVER as "5% is performance-based". - Ceiling is 25%. Jeff will NOT go above; he needs equity for CEO/CTO/employees. - Vesting stops ONLY on Abandonment (resignation, or 90 days no work + no communication after notice + 30-day cure). Disengagement short of that hits her GUARANTEED PAYMENT, not her equity. Two dials, never touch. - Repurchase right on bad leaver, but never on the first 10%. REALITY CHECK (Jeff, correcting Claude): Sol has ~1,000 IG followers and NO audience. She does not want to be the face beyond ~6 months. She is strong at social/branding/editing/on-camera/interviewing and speaks to the customer better than Jeff. She will NOT be the biggest creator account. Jeff is the engine (app, vision, AI marketing, facilitator network, in-person movement building, the buck stops with him). 25% is priced for the partnership, not for forward contribution. Real risk = her frustration/drift, and Jeff wants a dignified off-ramp for EITHER founder. SOCIAL ACCOUNTS: Company property, registered social@affirmology.ai. Credentials currently ONLY on Sol's phone -> MUST move into a company vault. Her authority is contingent on Active Service and reverts on Abandonment. ENTITY ARCHITECTURE (new): Affirmology (app/subscriptions) | Sacred Synergy (Jeff+Colin: retreats, 1-day events, programs, webinars) | Jeff Parker Love (personal brand). Line = Affirmology sells a tool, Sacred Synergy sells a room. Sacred Synergy PAYS Affirmology a referral share on sourced attendees; SS + JPL enroll as STANDARD creator/affiliate accounts (no sweetheart terms). Customer data NEVER transfers (Affirmology markets partner offers to its own list). Related-party deals need SOL's approval (Jeff+Colin both conflicted). Colin recuses on all SS matters. BUSINESS PLAN v6: retreats REMOVED as an Affirmology line (was a corporate-opportunity conflict); now "partner-led experiences + referral share". Practitioner/certification platform STAYS inside Affirmology (software/licensing). OPEN: Colin voting seat w/ recusal vs non-voting advisor?; Colin's own ~5%; vesting start date; 83(b) vs profits interest (30-DAY CLOCK FROM SIGNING); agree the 5 ladder numbers WITH Sol; tax-distribution rate; governing law.

  ---

  2026-07-12 (later) - SOL EQUITY REVERTED TO PURE TIME. COLIN 2%. SECURITIES FLAG.

  • SOL: Growth Ladder KILLED. Jeff already told her time-based and the NUMBER matters to her; reneging would cost more trust than it could buy. FINAL: 10% at signing + 5%/yr => 25% at year 3, PURE TIME, no performance test anywhere.
  • Motivation moved OFF the cap table: (a) AUDIENCE GROWTH BONUS in CASH ($2.5K@10K, $5K@25K, $10K@50K, $25K@100K combined organic followers on company IG+TikTok; max ~$42.5K, vs 5% equity worth $1M at a $20M company); (b) Company FUNDS Sol's platform (editor, social manager, production, content budget) - she keeps her personal brand, company gets the audience; (c) comp escalator tied to revenue.
  • COLIN: has been MIA, never completed the demo, only brought Norm (whom Jeff already knew). NOT 5%. Offer = 2% vesting 24mo/3mo cliff, FROM JEFF'S UNITS (no Sol dilution), NON-VOTING advisory (he's Jeff's Sacred Synergy partner so a voting seat = permanent recusal). Earnable beyond 2% ONLY for non-capital doors (partnerships, venues, key hires).
  • SECURITIES WARNINGS (urgent, flagged to Jeff): (1) LLC units = securities; need Reg D exemption + Form D within 15 days of first sale + state notices. (2) Sending an offering doc to 40-50 people without pre-existing relationships = GENERAL SOLICITATION, breaks 506(b), can create rescission rights. Use 506(c) (verified accredited) if marketing broadly. (3) NEVER pay Colin (or anyone) equity for INTRODUCING INVESTORS = unregistered broker risk. Get a securities attorney BEFORE Norm's check clears.
  • EMAIL LIST: CANNOT sell/transfer to Sacred Synergy (CCPA "sale" for valuable consideration; affiliate exemption needs common control AND common branding - SS has neither; as of Jan 2026 a sale/share also triggers a mandatory risk assessment). CAN: Affirmology emails its OWN members about an SS event as an Affirmology promotion, SS pays for the placement, data never leaves. Custom-audience ads with own list OK but = "sharing" under CPRA -> needs privacy policy disclosure + opt-out + Do Not Sell/Share link. Birth data + spiritual practice edges toward SENSITIVE personal info. ACTION: write the privacy policy BEFORE launch.
  • SACRED SYNERGY: not yet an LLC; form it before any referral deal. Money flows ONE WAY: SS PAYS Affirmology for promotion/sponsorship/placement. Affirmology pays SS nothing (Jeff doesn't want creator payouts). Boundary refined: Affirmology = community calls/workshops/3-4 hr activations (marketing+retention). SS = full-day+ paid retreats/programs.
  • BOARD: keep voting board tiny (Jeff + Sol + at most 1 independent). Build a NON-VOTING ADVISORY BOARD for Colin, Norm, and future risers (Monica Arenas, Glo, Lauren). Elizabeth Moss unresponsive. Jacque comes with the CEO conversation.
  • DILUTION: Sol must be told her 25% is PRE-investment and dilutes pro-rata (e.g. $50K at $1.5M post => Sol ~24.2%).

  ---

  2026-07-12 (late) - SECURITIES: THE REAL PICTURE + PLAN TO NORM'S CHECK (JULY 23)

  CORRECTION TO EARLIER ADVICE (I was wrong, verified against eCFR): - The "audited balance sheet for non-accredited investors" rule is DEAD. Repealed effective 2021-03-15. Current Rule 502(b)(2)(i)(B): a 506(b) offering up to $20M gives non-accredited investors UNAUDITED GAAP financials (Reg A Tier 1 / Form 1-A Part F/S(b) standard). NO AUDIT. Most of the internet still repeats the old $2M/120-day/audited rule. The REAL cost of non-accredited investors is a DISCLOSURE DOCUMENT (PPM), ~$5K-$20K from counsel - or ~$500-$2,000 if we DRAFT it and only pay for REVIEW. - The March 2025 SEC 506(c) verification relief does NOT help: it requires a $200,000 minimum per person. URGENT / #1 RISK: There is a VIDEO ON A PUBLIC LANDING PAGE (results page) saying we are raising money. "Unrestricted public websites" is the SEC's own named example of GENERAL SOLICITATION. This breaks 506(b). TAKE IT DOWN. Also sweep social, newsletter archive, deck links, any video. Also: 2 unknown people got past the gate and may have seen the offering doc -> do NOT sell to them under 506(b). GOOD NEWS: The exemption condition is broken by an OFFER, but LIABILITY (Section 12(a)(1) rescission) attaches on a SALE. NOTHING HAS BEEN SOLD YET. Fix before Norm wires and this is a non-event. Rule 508 CANNOT cure a general-solicitation violation. Waiting 30 days does NOT launder it (Rule 152(a)(1) still requires investor-by-investor proof they weren't solicited by the public communication). DECISIONS: Path = Rule 506(b), private, NO public mention of the raise anywhere for the whole offering. Jeff will DIY the disclosure document (SEC Form 1-A Part II checklist) + adversarial review in a second Claude chat + pay $500-$2,000 for securities-attorney review. He needs most of the doc ANYWAY because ANTIFRAUD APPLIES TO ACCREDITED INVESTORS TOO. DILUTION (answered): Investors MUST dilute both founders pro-rata. If units came only from Jeff's stake it would be a SECONDARY sale and the money would go to JEFF, not the company. Do NOT anti-dilute Sol: it compounds (CEO + CTO + employees + Round 2 would all hit Jeff alone), it looks like a red flag to investors, and dilution is not a loss (22.5% of a funded company >> 25% of a stalled one). Give Sol PRO-RATA RIGHTS instead, plus she already holds a Major Decisions veto on admitting new members. KEEP the existing promise: employee/CEO grants come from JEFF'S units, investor dilution is shared. CRITICAL PATH (the real bottleneck is NOT securities): can't take money without a BANK ACCOUNT -> can't open bank without a SIGNED OPERATING AGREEMENT -> can't sign without SOL -> can't sign without deciding 83(b) vs PROFITS INTEREST first (30-day clock starts at signature, cannot be reopened). FOUNDING SPONSOR IDEA (grounded, per Jeff): NOT a retreat-software play (nobody has 200-person retreats, nobody pays $5K for software at a retreat). The right analogy is CONFERENCE SPONSORSHIP (Aurea): pay now for privileged access, featured placement, a held category/region, inclusion in our in-person experiences, at a founding rate locked below anything we'll offer again. Creator accounts are a PAID tool tier (creators pay + burn credits), so "free creator account" was never a real offer. Mio is likely NOT the right first sponsor (Greece-based, small list). Sell the program to whom it genuinely serves. LEGAL LINE: a sponsor/partner may earn from THEIR OWN referrals (their effort, not a security). They may NEVER get a share of Affirmology's profits/revenue (passive return from Jeff's effort = security under Howey). NEW DOCS: Affirmology_RaisingMoney_FullReport_v1.docx (the full map), Affirmology_PlanToNormsCheck_v1.docx. ACCOUNTABILITY: daily 8am scheduled check-in "affirmology-raise-checkin" running through July 23.

  ---

  2026-07-15 - OA v6 (Jeff's refinements) + Sol docs updated

  Changes in Affirmology_OperatingAgreement_DRAFT_v6.docx (and both Sol Google Docs): - VESTING NOW MONTHLY (~0.417%/mo), not annual cliffs. Leaving at 18 mo = 17.5% vested. (Jeff's ask.) - DISTRIBUTIONS paid on VESTED % (e.g., 20% in yr2), rising to 25% when fully vested. Unvested = no distribution. - NEW Section 5B "Granting equity to the team": compensatory equity = Rule 701 exempt (NOT investor process). Three tools: (1) membership units/capital interest = becomes Member; (2) profits interest = future-only upside, becomes Member, tax-advantaged, standard for CEO/key hires; (3) phantom/profit-share = contract, NOT a Member. Jeff can grant up to 1%/person and 5%/12mo from HIS units unilaterally; bigger or new Member = both founders. The 1%/5% cap explicitly covers phantom/economic equity too (so Sol can't be routed around). - CEO EQUITY SOURCE flexible: from Jeff's units so long as Jeff stays >= [60%]; below that, excess via new-issuance dilution of all, by mutual agreement (CEO benefits both). [60% floor is OPEN.] - BONUS AFFORDABILITY GATE: paid only from operating cash flow, only after 6-mo reserve, installments allowed; OPTION to gate the two biggest tiers ($50K@100K, $100K@250K followers) on a min active paying-member count. - UNITS: 100,000 total (Jeff 75,000 / Sol 10,000 + 15,000 reserve). Recommended over 1,000 for whole-unit small grants. - Clarified this IS a real signable OA (can sign to open bank; counsel formalizes later). MIO CORRECTION (Jeff was right): now that non-accredited needs NO audit, Mio CAN invest under 506(b) as a non-accredited SOPHISTICATED investor IF (a) no general solicitation reached her (NDA-gated demo, not public), and (b) she gets the disclosure package (being built anyway) + documented sophistication. Straight membership units only, NOT a profit share (profit share to a passive investor = security AND messier). The real question is worth-it, not can-she. NEW Google Docs for Sol: OA v6 = https://docs.google.com/document/d/1UFfdGaQK_DMs8JO5-oLO9lH8HbNnpqD3XQi3JUt1bgA/edit Explainer v2 = https://docs.google.com/document/d/1MWI4JDRNO-i2Ltm92t9N3cn6FgPK2oKDVY_kwYVVs_I/edit STILL OPEN: 83(b) vs profits interest (30-day clock); vesting start date; [60%] CEO floor; bonus member-count option; tax-dist rate; governing law; 3rd board seat; Colin final; Faena signature; EDGAR; disclosure-doc financials+counsel.

  ---

  2026-07-16 - LAUNCH TIMELINE RESET (electional + practical, committed)

  Stop the start-stop. ONE date, do not move it again. Grounded in the agent council (Hermes) reading of Jeff's natal chart + web-verified astronomy (Mercury direct July 23; shadow clears ~Aug 11; total solar eclipse Aug 12 Leo on Jeff's Vertex; partial lunar eclipse Aug 27-28 Pisces on Jeff's 2nd-house money point). - Now - July 30: retreats + event (retreat to ~7/22, event 7/26, couples retreat 7/27-30). Do NOT touch launch. In pockets: lock Norm's money, buy sound beds, harden app, prep community message. - Aug 8 (Lions Gate): BETA opens to inner orbit + public TEASER/waitlist post. Not the formal launch (Mercury still in shadow). - Aug 12 (solar eclipse on Jeff's Vertex): NOT a launch. Fated-encounter energy for significant conversations. (NORM SIGNING IS JULY 23, not Aug 12 - the day Mercury goes direct.) - Aug 14: PUBLIC APP STORE LAUNCH + funnel + charging. Venus-Jupiter Leo money window; Jeff natal Venus tied to MC. NAIL THIS. - Aug 27-28 (lunar eclipse, 2nd house): CLOSE investor conversations already open (Norm commits). - Sept 11 + Faena Sept 24-26: in-person hard debut + community push (unchanged). SPECIAL: "Eclipse Portal" beta gift - limited Leo-eclipse identity-reset audio suite + Origin Soul Song for founding beta, framed as catching a rare cosmic window. Only offered to the founding group. Deliverables: Affirmology_LaunchTiming_August_v1.docx + a community "in development / founding circle" message. NOTE: Hermes council reached via cloud Studio API (https://affirmology-studio-api.onrender.com, health ok, corpus ok).

  2026-07-16 DECISIONS (OA v13 lock + bank/Norm sequencing)

  • SOL EQUITY = PROFITS INTEREST, LOCKED (Jeff's call). Reason: a priced round (Norm, ~$1.5M) is expected within days of signing, and a profits interest keeps her grant-date value at $0 regardless of that round, so no tax at grant or as it vests (Rev Proc 93-27). Threshold set at grant-date fair value (expected nominal) by the accountant. Protective 83(b) filed within 30 days of signing anyway (no IRS fee, no tax owed).
  • FULL OWNERSHIP SPELLED OUT: OA §2/§3/§11 and the ForSol doc now explicitly state a profits interest = full ownership from day one: Member, votes on all Major Decisions, ordinary pro-rata distributions on vested %, pro-rata participation rights, full upside above nominal threshold, 10% permanent floor, to 25% on time alone. "No doubt about her full profits interest situation" per Jeff.
  • Also locked earlier same day: Colin = interim tie-breaker until permanent independent board seat (Elizabeth Moss target); Amendment = anytime by both founders' written agreement (material also Board); Separate role/pay agreements clause added.
  • FILES: Affirmology_OperatingAgreement_DRAFT_v13.docx (signable), Affirmology_ForSol_Everything_v2.docx, Affirmology_BankAndNormSequencing_v1.docx.
  • TIMELINE (Jeff): leaves 5pm 7/16; returns eve of 7/22 to firm + sign; bank AM 7/23; Norm ~noon 7/23.
  • BANK does NOT gate anything. Bring: EIN letter (42-3770990), filed Articles (2026-002015324), signed OA v13, photo IDs + SSNs for both, optional Good Standing cert, initial deposit. Traditional bank same-day; Mercury/Relay online but 1-2 day verify.
  • SEC: Form D due 15 days AFTER first sale (Norm's commitment), no bank account needed to file. START EDGAR Form ID + Login.gov NOW (2-4 business day processing) so credentials are ready. Send Norm the disclosure packet anytime; his commit starts the 15-day clock, not before.
  • 83(b) is the only hard clock: 30 days from signing (~Aug 21 if signed 7/22). Grant + file BEFORE Norm funds.

  2026-07-16 PAY-PROTECTION FIX (OA v14, ForSol Everything v3)

  • Sol/Jeff flagged the real gap: v13 §5 let the Manager "reduce or suspend" pay at pure discretion "with notice."
  • FIXED in v14 §5: pay reductions now require a genuine, MATERIAL, DOCUMENTED drop in the founder's actual engagement, good faith, written notice, tied to reduced-role/Abandonment process, NOT Manager discretion; disputes go to mediation BEFORE any reduction takes effect.
  • §12 added "Once set in writing, pay is locked for its term": when revenue supports a specific pay level and the founders write it down, that level is binding for its term, not unilaterally reducible; changes only by mutual agreement or for cause.
  • §12A strengthened "Separate role and pay agreements, binding for their term" + new "Pay is set by agreement, never by fiat" clause: no fixed salary now because no revenue (feature, not loophole); founders get MORE protection than outside hires (who get narrow in-scope agreements); founder pay once written locks for term.
  • ForSol Everything v3 mirrors all of this in plain English (three new lead paras after the pay section).
  • Framing Jeff wants preserved: the OA protects founders from OUTSIDERS' loopholes; founders have more protection; he cannot screw Sol; looseness now is only about unknown numbers, not permission to cut.

  2026-07-16 83(b) CORRECTION (OA v15, ForSol Everything v4)

  • Jeff correct: with a PROFITS INTEREST, NO 83(b) election is required. Confirmed via Rev Proc 93-27 (no tax at grant) + Rev Proc 2001-43 (unvested profits interest, no election needed if Company treats her as Member from grant + she reports her share). Removed the "hard 30-day clock / unmissable" framing from §13, §18, and ForSol doc. Optional protective filing = accountant's call only, NOT required.
  • No hard deadline hangs over Sol's equity anymore. Only remaining action: accountant confirms structure
  • sets distribution threshold at grant-date (nominal) value; grant before Norm funds.
  • Current signable set: Affirmology_OperatingAgreement_DRAFT_v15.docx, Affirmology_ForSol_Everything_v4.docx.

  2026-07-16 FOUNDER PAY = CAPITAL-FUNDED RUNWAY (CompLetter No.1 v2, OA v16)

  • Jeff: founders WILL pay themselves from Norm's capital. $8k/mo combined Founder Support, funded for 2 months under the $30k raise (~$16k) and 3 months under the $50k raise (~$24k), matching the prospectus use-of-funds. After the funded runway, the revenue trigger carries pay (trailing-3-mo revenue covers Founder Support + 1 mo reserve buffer), capped 35-40% trailing-3-mo revenue; step-ups via signed letters.
  • Updated: Affirmology_FounderCompLetter_No1_v2.docx and OA v16 §12 (founder-support line now says support is paid from raised capital for a disclosed runway, then continues from revenue). Six Principles v1 still current.
  • Current signable set: OA v16, ForSol Everything v4, FounderCompLetter No1 v2, SixPrinciples v1, BankAndNorm v1.

  2026-07-16 PERSONAL-BRAND BOUNDARY + NON-COMPETE + GOVERNING LAW (OA v17)

  • Jeff's pressing blocker: draw a CRYSTAL-CLEAR line between Sol Ballard's personal brand (hers) and Affirmology brand work (company's). The old §15 IP line "Sol's brand and content work" was dangerously broad and could have swept her own name in. FIXED.
  • NEW §9A "The clean line: Sol Ballard is Sol Ballard, Affirmology is Affirmology": defines what's assigned (app, engine, corpus, Company accounts + every post to them, work created FOR Affirmology) vs what's HERS and never assigned/restricted (name, likeness, personal accounts, coaching/speaking/writing/courses, her pre-Affirmology "be uniquely you"/magnetic-on-camera body of work, future personal content). Overlap rules (company builds its OWN face-independent content engine so it survives without her; while serving she can be the face of Affirmology posts (company's); she can post her journey on her own accounts (hers); collabs each keep their own copy; write a content map). No deletion/destruction on exit.
  • CHART "MAGNETIC" IDEA handled in §9A: Affirmology's core = subconscious transformation via audio, NOT chart-combining. Sol may build a chart-based personal-brand report; inside app = Affirmology flavor driving to her programs; on exit she can take the branch via engine license OR build her own on the general idea, as long as not a copycat personalized-affirmation-AUDIO product and no member data/corpus travels. Option to contract it as a Sol-to-Affirmology license from the outset.
  • §15 NON-COMPETE narrowed: 12 months, ONLY a substantially-similar AI-personalized cosmic-blueprint affirmation/subconscious AUDIO product. Explicitly does NOT touch her personal brand, coaching, media, retreats, books, or a non-audio chart offering. Non-solicit 24mo but doesn't stop her talking to her own audience.
  • §17 GOVERNING LAW set to FLORIDA, venue Miami-Dade (convenience). FLAG (verified 2026): Florida (2025 CHOICE Act) = most employer-friendly, strongly enforces non-competes; Wyoming (eff. 7/1/2025) VOIDS most non-competes. So Florida makes §15 MORE enforceable -> keep §15 narrow to protect Sol regardless of forum. Counsel to confirm.
  • Sol-facing plain doc: Affirmology_ForSol_YourNameIsYours_v1.docx.
  • Current signing set: OA v17, ForSol Everything v4, ForSol YourNameIsYours v1, FounderCompLetter No1 v2, SixPrinciples v1, BankAndNorm v1.

  2026-07-16 SOL ROLE-BREADTH + GRAND SUMMARY (OA v18, Everything v5, CurrentSet index)

  • Sol's concern: on paper her role looked limited to social/followers, and the bonus now depends on funnel conversion (beyond her control). She's fine leaving the bonus (it doesn't drive her) but wanted it clear she's a founder whose value spans events, event strategy, ideas/direction, voice-of-customer, and in-person member acquisition, compensated broadly (salary + ownership + distributions), NOT per-lead.
  • OA v18: §4 "Scope of this bonus, and what it is NOT" (optional extra, not per-lead, not a measure of worth); §5 "The role, in full" (names events/strategy/in-person acquisition; much organic/in-person membership comes through her events not just followers; paid as a founder for all of it).
  • ForSol Everything v5 mirrors this in plain English.
  • GRAND SUMMARY built: Affirmology_CurrentSet_GrandSummary_v1.docx lists the canonical set + decided vs open
  • old files to ignore. Per Jeff: keep ONLY the current-set files, discard old versions.
  • CANONICAL CURRENT SET: OA v18; FounderCompLetter No1 v2; SixPrinciples v1; ForSol Everything v5; ForSol YourNameIsYours v1; InvestorPacket Norm v1; DisclosureDocument DRAFT v1; NormAdams UseOfFunds v2 (RECONCILE founder-pay line: letter=$16k/2mo@30k, $24k/3mo@50k vs UoF v2 $16k both + $7k extended); TermSheet v2; BankAndNormSequencing v1; Faena MC Agreement Signable v1; ColinAdvisor TermSheet v1.
  • OPEN for the 22nd: vesting start date; bonus milestones (keep/adjust/drop); UoF founder-pay reconciliation; tax-distribution rate (40%) + spend cap band (35-40%); Colin seat/grant; counsel review non-compete + 9/9A; start EDGAR credentials.

  2026-07-17 RAISE CHECK-IN (automated, 6 days to Norm)

  • VERIFIED LIVE: demo.affirmology.ai/result STILL publicly serves the raise. Confirmed by fetching the live page: "Investor Opportunities / Own a piece of Affirmology / from as little as $5,000" CTA block, plus the hoisted mission film ("how you can be part of it") directly above it. In repo affirmology-site/result.html (~lines 745-852). Also agreement.html line ~189 mentions joining "as an investor". This is the #1 critical-path item (general solicitation vs 506(b)); flagged 7/14, unchanged as of this run. Removal is a small edit + push to master (CI auto-deploys since 7/16).
  • EDGAR credentials still not started (listed OPEN for the 22nd; 2-4 business day processing).
  • Check-in delivered to Jeff with a ready paste block for the code chat to strip all public raise mentions.
  • SECOND RUN SAME DAY: re-fetched demo.affirmology.ai/result live; the "Investor Opportunities / Own a piece of Affirmology / from as little as $5,000" block is STILL being served (page lines ~832-854 as rendered). EDGAR still not started. Paste block re-delivered. 6 days to Norm (7/23).

  2026-07-18 RAISE CHECK-IN (automated, 5 days to Norm)

  • VERIFIED LIVE AGAIN: demo.affirmology.ai/result still serves the raise ("Investor Opportunities / Own a piece of Affirmology / from as little as $5,000", rendered lines ~835-854, plus "as an investor" mentions at ~832 and ~854). Third consecutive check-in flagging it (7/14, 7/17 x2, now 7/18). Unchanged.
  • EDGAR credentials: no evidence started (was OPEN as of 7/17; 2-4 business day processing means starting today/Monday is the last comfortable window before a 7/23 first sale -> Form D due ~Aug 7).
  • Jeff traveling (left 7/16, back eve of 7/22). Check-in delivered with the code-chat paste block re-issued and a note that EDGAR Form ID + Login.gov can be started remotely from a phone/laptop.

  2026-07-19 RAISE CHECK-IN (automated, 4 days to Norm)

  • VERIFIED LIVE AGAIN (fetched today): demo.affirmology.ai/result still serves "Investor Opportunities / Own a piece of Affirmology / from as little as $5,000" (rendered lines ~837-839) plus "as an investor" at ~832. FOURTH consecutive check-in flagging it (7/14, 7/17 x2, 7/18, now 7/19). Unchanged.
  • EDGAR credentials: still no evidence started. Today is Sunday; Monday 7/20 is the LAST realistic start for 2-4 business day processing ahead of a 7/23 first sale (Form D due ~Aug 7 either way).
  • Jeff traveling, back eve of 7/22; sign 7/22, bank AM 7/23, Norm ~noon 7/23. Check-in delivered with the code-chat paste block re-issued; both the site fix and EDGAR are doable remotely from the road.

  2026-07-20 RAISE CHECK-IN (automated, 3 days to Norm)

  • VERIFIED LIVE AGAIN (fetched today): demo.affirmology.ai/result still serves "Investor Opportunities / Own a piece of Affirmology / from as little as $5,000" (rendered lines ~837-839) plus "as an investor" at ~832. FIFTH consecutive check-in flagging it (7/14, 7/17 x2, 7/18, 7/19, now 7/20). Unchanged.
  • EDGAR credentials: still no evidence started. TODAY (Mon 7/20) is the last realistic start for 2-4 business day processing ahead of a 7/23 first sale (Form D due ~Aug 7 regardless).
  • Jeff back eve of 7/22; sign 7/22, bank AM 7/23, Norm ~noon 7/23. Check-in delivered; code-chat paste block re-issued. Both items doable remotely from the road.

  2026-07-20 (eve) NORM DOC SET REBUILT AS GOOGLE DOCS + TRASH RECOVERY (Cowork legal/investor chat)

  • WORKFLOW NOW LOCKED: all investor deliverables are NATIVE GOOGLE DOCS created via the Drive connector (HTML, Georgia/emerald headings + gold rules, Arial body) in Drive folder "01 - For Norm (July 23)" (id 1tzDs-LtCnF4OsepmbKBMcE9Vvm3B5QsB). NO .docx deliverables ever again (MEMORY.md 2026-07-20).
  • BUSINESS PLAN v10 pushed to the Norm folder as a Google Doc. Adds, per Jeff's decisions: (a) the full 3-tier credit/referral economy from AffiliateRewards_Model_v1 (member credits + comped months; Creator Accounts ~10 referrals with status/access/comped premium; paid influencers 25-35% rev-share capped 12 mo), every number marked "indicative, finalized from beta"; (b) viral/share mechanics section (shareable video cards, gift-a-Soul-Song LIVE AT LAUNCH with in-app-only redemption so gifting = acquisition + both sides earn credits, event video cards, live infra: OG cards + watch pages + Mirror drip); (c) Golden Tier "companion that remembers" narrative from PersonalKnowledgeBase_GoldenTier_Concept_v1 (credits = what bespoke creations cost, allotments, no rollover, never throttle creators); (d) chart-informed marketing (Lion's Gate 8/8 beta, Eclipse Portal, Sept 11 launch kept as in v9); (e) OA-v18 governance (profits interest full-ownership language, 3-seat board, Colin 2% non-voting interim tie-breaker, pay protection), flagged "a few points finalize Wednesday"; (f) flexibility language throughout (tech help may become a $30-40K contract or team hire; audio beds may become an audio engineer at ~1% equity; everything recalibrates on beta). Source HTML saved as Affirmology_BusinessPlan_v10.html in the project folder.
  • DISCLOSURE v3 pushed as a Google Doc. Section 4 now carries the exact $50K-column figures per Jeff: founder 40%, ads 13%, engineering 10%, events+venue+demo+oracle 15.4%, beds+API 11.6%, hardware/creative 6%, accounting+buffer 4% (sums to 100; strict line math is events 15.0 + acct/buffer 4.4, Jeff's split assigns the $200 delta to events). Also kept the Drive-v2 improvements (profits-interest ownership row, promissory-note terms, founding-investor bonus disclosure). Jeff to trash DRAFT_v2.
  • TRASH RECOVERY: Jeff's old-version cleanup had accidentally trashed the CURRENT TermSheet_v3 (bonus ladder, $15K circle, Coherence partnership) and InvestmentElection_v2. Both RESTORED from Drive trash via Chrome back into the Norm folder and verified by content. Still to trash (Jeff): TermSheet_v2, InvestmentElection_v1, DisclosureDocument_DRAFT_v2, BusinessPlan_v7.
  • MISSION DOC: Contents page added. Built from Affirmology_Mission_and_Opportunity_PRINT.html with ONLY a new "The map of the story" ToC page inserted after "A note on how to hold this" (existing type system: mono gold part labels, Cormorant titles, dotted gold leaders, real printed page numbers measured from the 61-page render; nothing else touched). REVIEW COPIES in the project folder: Affirmology_Mission_and_Opportunity_v8_TOC_REVIEW.pdf + Affirmology_Mission_and_Opportunity_PRINT_v8toc.html. NOT pushed to R2; Jeff reviews first, and his feedback becomes the v8 work order (v7 corrections proposal in Affirmology_PROPOSAL_MissionDoc_v7_Corrections.md still awaits his go).
  • Norm folder verified current set: BusinessPlan_v10, DisclosureDocument_DRAFT_v3, TermSheet_v3, NormAdams_UseOfFunds_v5, NormAdams_InvestmentElection_v2, InvestorPacket_Norm_v2, DemoExperience_v5.

  2026-07-21 (overnight) - NORM DOC SET FINALIZED (v11/v4/v7/v3/v3/v4) + MISSION DOC REVIEW r4 (Cowork chat)

  • DRIVE FOLDER "01 - For Norm (July 23)" IS THE FINAL SET, 7 native Google Docs, house format, old versions trashed: BusinessPlan_v11 (18G79CyFqr...) - version line trimmed to "v11 · July 2026. Figures marked indicative are finalized from beta data." (internal changelog inventory removed per Jeff); DisclosureDocument_v4 (1yl_gDDMp...) - use-of-proceeds re-synced to UoF v7: founder 40 / ads 13 / engineering 10 / events 15 / beds+API 11.6 / hardware+creative 8 / accounting+buffer 2.4 (=100); NormAdams_UseOfFunds_v7 (1Iw7jlZ_...) - Jeff's new line order: audio beds 4000, API 1800, oracle decks 1000, demo gear 5500, video 1000, graphics 500, Sept-11 venue 1000, Sol laptop 1000, MAC MINI NOW $1,500, tech help 5000, accounting 300, founder 20000, ads 6500, buffer $900 (=50,000 exactly; $30K col =30,000). "Extra $20K" section reordered tech-support first, then ads, then demo; founder-focus bullet REMOVED. v5 and v6 and the interim v7 push all trashed; InvestorPacket_Norman_v3 (1fOTRj3k...) - page-2 orphan fixed (checkboxes 9pt + Part-1 field spacing 12->8pt); 5pp, each part opens its own page; InvestmentElection_v3, TermSheet_v4, DemoExperience_v5 unchanged.
  • Norman share blurb (per-doc one-liners + how-to-prepare) delivered in chat; Jeff pastes it with the folder link (normadams@comcast.net, read+comment; Jeff sets the share).
  • MISSION DOC REVIEW r4 (72pp) committed to this folder (REVIEW_r4.pdf + PRINT_v8.html source). NOT on R2, NOT in Norm's folder. Fixes to Jeff's r2 complaints: true full-bleed print pass (CSS named pages: cover + featurepages @page full margin 0; body @page 0.92/0.98/0.78in), all CSS page frames disabled in flow and replaced by post-render reportlab stamp (0.75pt gold frame inset 0.3in + centered page number, body pages only) so the frame can never cross text; ToC full-page, numbers re-measured (note3 exec4 vision8 problem14 method16 science20 product26 stay29 data31 market39 gtm42 rollout50 team53 state58 moat60 ask63 risks67 future70 invite72; Why-People-Stay row now listed before Data to match body order); duplicate "repetition wiring" figure removed; demo-lounge photo moved into GTM beside the spearhead paragraph; Huxley + "Not one audio" quotes merged into their pages; Atlas screenshot re-cropped (baked caption + cream margins stripped) and enlarged; lone-image pages set as centered plates. Residual airiness only on team wrap-up + state checklist tails (flagged as optional fills). Pipeline: WIP_v8.html -> RENDER_v8.html (local @fontsource) -> chromium print-to-pdf -> corner-pixel page classify -> reportlab stamp overlay.
  • Mac folder synced: BusinessPlan_v11.html, DisclosureDocument_v4.html, InvestorPacket_Norman_v3.html, NormAdams_InvestmentElection_v3.html, TermSheet_v4.html, NormAdams_UseOfFunds_v7.html, Mission PRINT_v8.html + REVIEW_r4.pdf.
  • OPEN: Jeff reviewing mission r4 (R2 push only on explicit go). EDGAR + demo-site investor-language items from the raise check-ins remain open (other lane).

  2026-07-21 (later) - MISSION DOC r5: Jeff's page-by-page order applied (63pp, was 72)

  • Removed AI images per Jeff (human element + flow): system-and-soul braided plate, synchronized group journey, founding-chart plate (redundant), five-rings moat plate. Cosmos-opening plate moved to the BOTTOM OF THE ToC page. Ladder-of-doors plate moved to the State-of-the-Project checklist page.
  • Merges (killed lonely image pages): Jeff-in-room photo+Meade quote up to the vision page; community photo (caption dropped, 48%) onto the method page; constellation plate (cropped 10/10, no caption, 60%) onto the Constellation Effect page; couple plate (54%) onto the "Yes, they are a couple" page; walking-into-light (58%) onto the Future text page; "Upsell as depth" kbox joined its page by shrinking the day-with-Affirmology suite figure to 70%.
  • New/replaced images: hotline photo replaced with ORIGINAL Brand/photos/IMG_9238.jpg (was a blurry live-PDF crop) and placed with the product-ladder kbox; invitation couple photo replaced with ORIGINAL Brand/photos/FullSizeRender.jpg and restored to side-by-side text+photo flex (no text was ever lost - the photo had been isolated by flow, not content); two IG new-moon screenshots (Jeff's uploads "astrology new mooon image 1/2.PNG", status bars cropped) added to the why-the-subscription-holds page, caption "The monthly sky, live in the feed"; data-section second photo widened to full width to match the flywheel image.
  • Jeff-at-computer photo NOT FOUND on disk (searched root + Brand + MoodBoard + VIDEO_BUILD); Atlas screenshot page stands alone for now - slot ready if the photo surfaces.
  • ToC re-measured for 63pp (note3 exec4 vision8 problem12 method14 science18 product24 stay27 data28 market35 gtm38 rollout45 team47 state51 moat53 ask55 risks59 future62 invite63). r5 PDF + PRINT_v8.html committed to this folder. Still NOT on R2.

  2026-07-21 - MISSION DOC v8 APPROVED ("good enough" - Jeff) - PUBLISH IN FLIGHT

  • Final build: r10 (61pp), through 5 feedback rounds (r5-r10): AI plates culled for human element; images traced to true originals in Brand/photos (plate-loft-listening, IMG_9238 hotline, IMG_2959 invite, Systems to Audio, Replacement sequence labeled pipeline); v6 group-journey + chaise images recovered per Jeff's request; GTM / killer-app / team / invitation sections rebuilt to match published v6 layouts; ToC page live-measured each build.
  • CANONICAL FILE: Affirmology_Mission_and_Opportunity.pdf (this folder, 7,731,752 bytes, pikepdf-optimized). Source: Affirmology_Mission_and_Opportunity_PRINT_v8.html. v6 kept for reference: Affirmology_Mission_and_Opportunity_v6_FINAL.pdf.
  • PUBLISH: Jeff pasting a block into the CODE chat -> wrangler puts docs/mission-and-opportunities-v6.pdf (backup, from local v6 file) then REPLACES docs/mission-and-opportunities.pdf (verify content-length 7731752).
  • NORM FOLDER: Jeff dragging the canonical PDF into Drive "01 - For Norm (July 23)" (Cowork's browser bridge cannot read the folder's files for upload). Folder becomes 8 docs once in.

  2026-07-21 RAISE CHECK-IN (automated, 2 days to Norm)

  • VERIFIED LIVE AGAIN (fetched today): demo.affirmology.ai/result still serves "Investor Opportunities / Own a piece of Affirmology / from as little as $5,000" (rendered lines ~837-839) plus "as an investor" at ~832. SIXTH consecutive check-in flagging it (7/14, 7/17 x2, 7/18, 7/19, 7/20, now 7/21). Unchanged.
  • EDGAR credentials: still no evidence started. The 2-4 business day window ahead of a 7/23 first sale is now gone; Form D is still fine (due ~Aug 7, 15 days after first sale) but Form ID + Login.gov must be submitted TODAY or tomorrow to have credentials in hand near the deadline.
  • Jeff back eve of 7/22; sign OA v18 7/22, bank AM 7/23, Norm ~noon 7/23. Check-in delivered; code-chat paste block re-issued. Both items doable remotely.

  2026-07-22 RAISE CHECK-IN (automated, 1 day to Norm)

  • VERIFIED LIVE AGAIN (fetched today): demo.affirmology.ai/result still serves "Investor Opportunities / Own a piece of Affirmology / from as little as $5,000" (~lines 835-839) plus "as an investor" (~832) and "a tester or an investor" (~854). SEVENTH consecutive check-in flagging it (7/14 through 7/22). This is the last check-in before Norm's check; if it is still up when he wires, the public-solicitation problem attaches to the actual sale, not a hypothetical one.
  • EDGAR credentials: still no evidence started. Credentials will NOT be in hand before a 7/23 first sale; Form D itself is still fine (due ~Aug 7). Form ID + Login.gov must go in today regardless.
  • Plan of record: sign OA v18 today 7/22 (83(b)/profits-interest 30-day clock starts at signature), bank AM 7/23, Norm ~noon 7/23. Check-in delivered; code-chat paste block re-issued.

  2026-07-23 RAISE CHECK-IN (automated, NORM DAY - 0 days)

  • VERIFIED LIVE AGAIN (fetched this morning): demo.affirmology.ai/result STILL serves "Investor Opportunities / Own a piece of Affirmology / from as little as $5,000" (~lines 835-839), "as an investor" (~832), "a tester or an investor" (~854). EIGHTH consecutive check-in flagging it, and today is the day of the sale itself. If Norm wires while this is up, the general-solicitation problem attaches to an ACTUAL sale under 506(b). Check-in led with this; code-chat removal paste block re-issued with instruction to take it down BEFORE the wire, this morning.
  • EDGAR credentials: still no evidence started. Cannot be in hand today; Form D still fine (due ~Aug 7 if first sale today). Form ID + Login.gov submission urged again today regardless.
  • Plan of record per 7/16: OA v18 signed eve of 7/22 (to confirm with Jeff), bank AM today, Norm ~noon. Reminded Jeff: accreditation questionnaire + subscription agreement signed BEFORE the wire.

  Related documents

  State of Affirmology: Week of June 20, 2026Affirmology Video Project, Handoff CapsuleCouncil Characters / Avatars - Project Brief (v1)Session State + Terminal Startup Guide (v1)