Home / Ops / More in this area

_SHARED_CONTEXT - the cross-chat shared brain (read me)

Updated Jul 13, 2026 · _SHARED_CONTEXT.md

_SHARED_CONTEXT - the cross-chat shared brain (read me)

A living, shared record for EVERY chat working on Affirmology: the DATA/engine chat, the code/apps chats, and Cowork. It is the "things worth knowing" layer, decisions, standing principles, current phase, and owner split, so we all stay aligned and Jeff does not have to re-explain across windows. Append to it as decisions are made. For minute-to-minute deploy coordination use _CHAT_SYNC.md; for the authoritative deep state use Affirmology_PROJECT_STATE.md. This file is the fast shared map.

Started 2026-07-04 at Jeff's request.

Current phase (as of 2026-07-04)

• Beta is NOT starting today. Target open: Fri 7/10 (Thu 7/9 tighter). Jeff on a family trip, back in the Miami office late Tue 7/8 / early Wed 7/9. Plan: Affirmology_BetaLaunch_RevisedPlan_v1.md.
• Gift drip to the list (done with care, not rushed): Mirror ~Sun 7/5, a new audio Mon/Tue, the report Wed/Thu. Founders excluded; maybe a few people trimmed.
• Today (7/4): a beta-expectations + affirmational-agreement email (Cowork owns the copy).
• The Mirror is HELD and being retooled (structure/content). Do not batch-send it.

Standing principles (do not violate)

• Map, never the full picture, for a creation prompt. Generation prompts get a compact MAP of reference material; the full detail is pointed to, never dumped in. Full stays checkable in the corpus + a FullReference doc. (First applied to the storytelling craft menu; bake it everywhere.)
• Iterate, don't regenerate. For a tweak, edit the person's EXISTING script and re-voice it verbatim (/render-verbatim). Only full-regenerate for a structural rewrite or with an explicit ask. Losing liked language to a re-render is the thing we are preventing.
• The Profile is the canonical source. Compute a person's chart once, store it, read from it for every audio/reading. Affirmology_ProfileCanonicalSource_v1.md.
• The council is our signature weapon AND our biggest cost. It is the expensive multi-round call (the Mirror uses it). Run it ONCE per person and cache the output; re-voice cheaply. Never fire it repeatedly at the same person.
• All audio comes from our system. Founders (Jeff, Sol) excluded from customer counts/reports. Never em dashes. "Subconscious Operating System" frame. Sol = Soledad Gabriela Ballard.

Cost discipline (learned the hard way 2026-07-03: credits hit zero mid-batch)

• Turn on Anthropic auto-reload; enforce a low-balance warning (not just a note).
• Trim prompts (the craft menu was 8.5k tokens/render, now a ~2.7k map, full on demand).
• No blind auto-retry loops (a keepalive re-firing a failing batch wasted cycles).
• Before any batch: dry-run + cost estimate + confirm.

Owner split

• DATA / engine chat (this one): the render/audio pipeline, profiles, chart engine, corpus/craft, cost-safe generation, the demo results page, gift renders, recovery.
• Code / apps chats: the iOS/Atlas apps, TestFlight, cloud-Fish, the Mirror batch runner, the app accounts and in-app library, the wiki.
• Cowork: the beta agreement text, outbound email copy, app/TestFlight account linking (Sol), wiki.

Decisions appended 2026-07-04 (Cowork session, reading-craft reset)

• THE UNDERSTANDING is decided (Jeff, 7/4). Code's "structured Understanding" = Cowork's "Deep Map"; one name: the Understanding. The council builds it ONCE per committed person (12-field schema v1 in Affirmology_FlagshipReading_Proposal_v1.md section 3: ranked salient findings, convergences, aha moves, wound/gift/gap, mechanisms, roads, metaphor bank, one-liners, fear preemptions, timing, context store, learned-info store). Every audio/chat after is a cheap shaping (select facets + apply craft). APPEND on new learnings, never recompute. Built INSIDE the Mirror rebuild. Two tiers: Tier 1 light (everyone gets Origin), Tier 2 the deep run on commitment (recommended trigger, pending Jeff: the "Yes, I'm in" beta reply).
• The hero's story is the product-wide standard. Every output makes the listener feel seen, epic, meaningful, and shown a path. Craft standard + Mirror v3 postmortem: Affirmology_ReadingCraft_HeroStory_Synthesis_v1.md.
• New first reading structure spec'd (public name stays The Mirror; internal mirror_recognition_v1): Promise, Being Seen, Three Doors (Big 3 convergences), Install, Seal; ONE threaded master symbol; a loop always open; 12-rule QC gate; test on Sol first. Affirmology_FirstReading_Structure_v1.md.
• Origin Soul Song B variant spec'd (origin_soul_song_b_craft, six line-level changes, separate axis from The Recognition A/B): Affirmology_OriginSoulSong_Bvariant_Spec_v1.md. Pronoun bug ("becoming her" in Jeff's gold script) must be fixed in both arms.
• The 72-min wealth reading transcripts were recovered (they existed only as MP3s on R2; Cowork transcribed both): Affirmology_WealthReading_Transcript_{Jeff,Sol}_v1.md, the canonical style exemplars. Known render flaws found there: composer meta-seams spoken aloud + Jeff's MP3 truncated at 71:37 -> QC now includes seam-strip and end-integrity checks.
• Beta invite email drafted (Affirmology_BetaInviteEmail_v2.html): craft-first framing, one small trip line, human-answers service line, reply "Yes, I'm in" as the affirmational signature. SEND DEPENDENCY: hello@affirmology.ai must receive replies (Email Routing forward) and the send must use a reply-able from, not the no-reply reports@ pattern.
• Jeff's fast map for launch week: Affirmology_LaunchWeek_OnePage_v1.md (his remaining open decisions: email send timing, tester list, Tier 2 trigger, starting audios set).

Decisions/state appended 2026-07-04 (render + cost-safety thread)

• Batch cost safeguards BUILT + DEPLOYED (affirmology-studio, commits 76840e8 + 6827996). POST /api/batch-mirror now enforces three independent brakes: (1) a spend cap (max_cost_usd, budgets ~$0.45/reading, stops BEFORE draining the pool below the demo's reserve), (2) stop-on-credit- error (aborts the whole run on the first Anthropic "credit too low" error instead of hammering), and (3) a gentle overnight throttle (pace_seconds, one-at-a-time drip; 25 over 8h = ~1140s apart). Workflow: Jeff states the balance before an overnight run, the cap is set to leave a demo reserve. Cost discipline is now ENFORCED in code, not just a principle. Batch stays HELD (all 27 in .mirror_hold) until the Mirror rebuild is validated.
• Hybrid render architecture spec'd (Affirmology_HybridRenderArchitecture_v1.md, DESIGN): cloud = always-on router + demo; local Mac = preferred render engine when awake, on its own credit pool. Cloud health-checks the local node and offloads renders to it BEFORE spending cloud credits, falls back to cloud when the Mac sleeps. Local overnight runner (caffeinate + idempotent). Jeff DEFERRED buying a second Anthropic account for now - running on the shared account with the safeguards above. Verified 7/4: local + cloud Anthropic keys are the SAME account (no isolation yet, so the safeguards matter).
• Audio-engine layered architecture (Affirmology_AudioEngine_LayeredArchitecture_v1.md) - companion to THE UNDERSTANDING decision; same model.
• Wealth transcripts also recovered by this thread (faster-whisper "small", same two filenames). If Cowork's transcription (with the seam/truncation QC findings) is higher-fidelity, treat Cowork's as canonical so there is one version.

• Site fully live (direct wrangler deploy; the Workers CI was stuck one commit behind): green Mirror player matching the Origin, NO download on either song, results page lands CENTERED on the Mirror. Investor PDF republished + live (media.affirmology.ai/docs/mission-and-opportunities.pdf).

• Beta acceptance mechanism spec'd (Cowork, 7/4): terms presented as email summary cards + a linked full agreement page (Affirmology_BetaAgreement_v1.md, page copy + internal tracking spec). Acceptance = the "Yes, I'm in" reply to hello@; logged in Airtable People (Beta committed + beta_affirmed_at + verbatim reply text -> the person's Understanding context_store). Every yes gets a confirmation email within the day (Affirmology_BetaConfirmationEmail_v1.html). Do NOT use the broken results-page checkbox for founding-circle acceptance. If Jeff approves the Tier 2 recommendation, the logging event also fires the overnight Understanding run.

• The Norm lesson (7/4, Cowork correction): grading people from CRM rows alone is WRONG; the Profile store is canonical (existing principle, now proven). Norm Adams "missing birth data" was only a thin CRM row; his full profile (Norman Walter Adams, Origin delivered 6/30, clean) existed all along. Cowork synced his CRM row. RULE: any people-triage must cross-check the profile store before declaring data missing; Group D of Affirmology_InnerOrbit_ListTriage_v1.md is provisional pending that cross-check (engine chat). Also: beta is named THE INNER ORBIT (customer-facing); invite email + agreement renamed, no headcount ever stated to the list.

Open threads to know

• Sol's app is empty because her audios are not attached to her app account (solballard@gmail.com, which exists). Backfill of her 7 audios is staged; waiting on her app-account id to attach them.
• Push test audios into Jeff's + Sol's app (in-app + email) via /api/library/ingest; needs the account-id link. See the beta plan doc.
• Storytelling/hooks craft now wired into script generation as a compact map; FullReference doc = Affirmology_StorytellingHooks_CraftMenu_FullReference_v1.md.
• Graceful failure shipped on the results page (a dead render no longer spins forever).

Pointers

• Deep state: Affirmology_PROJECT_STATE.md · Beta plan: Affirmology_BetaLaunch_RevisedPlan_v1.md
• Tactical deploy coordination: _CHAT_SYNC.md · Canonical profiles: Affirmology_ProfileCanonicalSource_v1.md
• Craft full reference: Affirmology_StorytellingHooks_CraftMenu_FullReference_v1.md
• Reading craft standard: Affirmology_ReadingCraft_HeroStory_Synthesis_v1.md · Reading structure: Affirmology_FirstReading_Structure_v1.md
• Understanding schema + tiers: Affirmology_FlagshipReading_Proposal_v1.md · Launch-week map: Affirmology_LaunchWeek_OnePage_v1.md
• Style exemplars: Affirmology_WealthReading_Transcript_{Jeff,Sol}_v1.md

Decisions/state appended 2026-07-04 (DATA chat: verbatim + recovery + graceful failure)

• Verbatim voice render shipped + live (POST /api/render-verbatim, self-contained module). Voices an EXACT edited script with NO regeneration, through the proven Origin pipeline, and persists a demo job so the permanent results page serves it. This is the enabler for the "iterate, don't regenerate" principle and for user-created variations later. Use it for every micro-edit (edit the stored script's targeted part, re-voice) instead of regenerating.
• Graceful failure handling shipped (affirmology-site, results page). The returning poller never checked for a failed render and never timed out, so a dead render (e.g. the credit outage) spun "Loading your page" forever with no email. Now: on a failed status or an 8-minute cap it shows a reassuring message; the song still emails when it lands. Covers both the emailed-link and first-time film paths. Any future dead render never strands a person.
• Chasity + Colin recovered (7/3-7/4). Both took the demo DURING the credit outage, so their Origin renders failed and they got nothing (no song, no email). After credits were restored, DATA re-rendered both from their stored profiles and emailed them their results pages. Chasity: chasitywill@gmail.com, Douglas GA. Colin still has TWO duplicate profile rows (cjacobsandassoc + colinconnection) to collapse.
• Craft menu = compact MAP, live (62688f2): storytelling/hooks wired into script generation and the council as a ~2.7k-token map (was 8.5k full); full detail on demand + ..._CraftMenu_FullReference_v1.md. Dedupe pass next. This is the first application of the map-never-full-picture safeguard.

Decisions/state appended 2026-07-04 (DATA chat: nurture ladder + cheap Mirror validated)

• The Nurture Ladder / Goody Menu is spec'd (Affirmology_NurtureLadder_GoodyMenu_v1.md). After the Origin Song, a person gets a short drip of DISTINCT, cheap, high-value goodies, each a new modality + angle, each opening a loop the gate closes: Rung1 Mirror Door One (~$0.50 spoken), Rung2 Signature Card (visual PDF, ~free), Rung3 a 2nd audio of a different type (~$0.20), Rung4 a living "Season" timing note (~$0.10). Whole ladder for a never-converter ~$1.00-1.50; full Understanding only fires on commitment. SAME shape serves both gates: pre-beta "Yes I'm in" now, PAID membership trigger post-launch. Sequencing: build the CHEAP Mirror + a couple goodies NOW for the pre-beta list; build the elaborate "$1-at-a-time" on-demand journey AFTER beta teaches us the real journey.
• The cheap one-door Mirror is VALIDATED at length. Sol's Mirror "Door One" chunk composed via /api/compose (single council pass, ~$0.50): 1,411 words, ~10.7 min paced, 0 em dashes, one loudest convergence (Gate 41: Sun+Venus Aquarius / HD gate 41 lines 2-3 / Gene Keys Life's Work + Venus Attraction + Pearl), full segment craft, clean open loop to "the second door." Saved Affirmology_SolMirror_Chunk1_TestScript_v1.md, emailed to Jeff for approval. NO audio rendered yet (awaiting Jeff's go, then Fish-paced render). This is the near-term pre-beta tool.

Decisions/state appended 2026-07-05 (DATA chat: Understanding cost VERIFIED, built for 4)

• The Understanding costs ~$0.20/person, VERIFIED across 4 people (was assumed $2-5). Single grounded Sonnet-4.6 pass over the stored Profile: Sol $0.20, Jeff $0.193, Staci Kessler $0.166 (all well under $0.50). Game-changer: the Understanding no longer has to be beta-only; it could go to far more people. At volume a gate still makes sense (~$200/1000) but it can be generous. Built via scratchpad build_understanding_multi.py with a NAME GUARD (contamination protection: Jeff's email resolves to multiple demo names incl. Josh Parini/Aurora Vale; guard confirmed "Jeff Parker Love" before building) and an INCOMPLETE-CHART GUARD.
• Understandings now exist for: Sol, Jeff, Staci Kessler. Jeff + Sol stored as pinned KB entries (as_person jeff/sol, auto-injected into future generation). Staci saved as a DOC only (no KB account for demo-takers). FOLLOW-UP NEEDED: wire per-profile Understanding storage so a demo-taker's Understanding actually powers their future readings (KB is only keyed to jeff/sol today).
• Marc Kessler BLOCKED: incomplete chart (sun=None, 0 chart keys in his stored profile). His demo submission never produced a full chart. Needs complete birth data (date, EXACT time, place) + a profile/chart rebuild before an Understanding or any audio. Hard rule held: no build from an incomplete chart.
• Sol Mirror render-ready (~12 min, her feedback applied) awaiting Jeff's render go. New STANDING CRAFT RULES from Sol's feedback: (1) NO wishy-washy/unsure language ever, be definite; (2) a little repetition ok, never say the same thing many times; (3) definite opening ("I took an in-depth look at your chart"), never "I've been sitting with your sky for a while"; (4) NO catastrophe/depressing framing ("everything is falling apart"), keep hope + power dominant.
• Membership reading-series longevity answered (Affirmology_MembershipReadingSeries_v1.md): the 10-12 min reading as membership heartbeat does NOT run out; chart is finite skeleton, but transits + life context + depth-spiral + combinations + lenses + narrative arc are infinite renewables. Year-1 arc: Revelation (mo 1-3) then Companionship (mo 4-12+).

Decisions/state appended 2026-07-05 (DATA chat: audios + Understanding storage wired)

• Understanding storage WIRED to the shared endpoint (thank you for building da7a66b). The generator now writes every person's Understanding via POST /api/profile-understanding (NOT the founder-only KB). Stored + verified reading back for Sol, Jeff (Jeff Parker Love / dcfa886a), Staci Kessler, Marc Kessler. NOTE: store founders by EMAIL (solballard@gmail.com / jeff@jeffparker.love), not by "sol"/"jeff" (bare account ids returned ok=False; email resolves to the Profile row). Understanding is stored as the full markdown doc string for now; can move to structured-dict later.
• Marc Kessler UNBLOCKED + built (your emailless-orphan dedup fixed his chart: Libra Sun, 6 systems). Understanding built $0.098, stored. All 4 test people now have Understandings on their profiles.
• Sol is fully done: Understanding + Mirror audio (Fish, 9:14) + Soul Song B 4min (4:22) + Soul Song B longer (6:10, carries the full 6:15 Heavenly Circuit bed). Jeff's set rendering now (Mirror + both Soul Songs).
• Mirror audio renders LOCALLY via Fish (mirror-male reference e686ae649...), not the cloud: a minimal venv on the Mac (PYTHONPATH=affirmology-studio/affirmology-agent/src, deps: requests python-dotenv pydantic pyswisseph geopy timezonefinder boto3) calls synthesize_voice_fish_paced directly, runs scripts/audio_qc.py, uploads to R2 media.affirmology.ai/mirror/. FUTURE IMPROVEMENT for whoever deploys the studio next: give /render-verbatim an optional structure_id (default origin_soul_song) so passing "the_mirror" renders the Fish voice cloud-side (pipeline already routes mirror-male -> Fish). I prototyped it but reverted to keep the repo clean around your WIP.
• Soul Song length calibration (for /render-verbatim origin renders): ~2.1 words/sec. 4-min-under-4:30 = ~530-560 words; full-6:15-bed = ~840-852 words. Gate the word count BEFORE rendering (model overshoots word targets), then render once and ffprobe to verify.

Decisions/state appended 2026-07-11 (Cowork briefing chat: beta action map + corpus rule)

• Affirmology_AudioPossibilities_Map_v2.md is the WORKING beta map (v1 stays as the palette reference). It contains the ranked Wave 1/2/3 build order (Wave 1 floor: the Daily Pair S5 Night Install + S11 Morning Claim as locked structures, plus the feedback instruments), new structures S20 The Belief Surgery + S21 The Identity Gym, the Habit Engine (the Update Install frame: SOS version numbers, audios as patches, "1 update pending" as the native loop; identity-named streaks; anchor-word carry system; locked ethics line), the Assimilation Engine (7-step loop; only rating capture + feedback-to-craft-confidence write-back unbuilt), and the testing kit. Owner split for Wave 1: DATA/Studio = structure templates + renders; Code = library-ingest journey delivery + 2-tap ratings in app; Cowork = daily email copy.
• NEW HARD RULE, all chats: never manually upload corpus/corpus.db to R2 without pulling the current one and checking LastModified first (cron pushes ~09:00 UTC); merge, never replace. On 7/11 a 09:36 UTC manual upload overwrote the cron's enriched db and wiped a fresh 42-entry somatic/attachment/IFS batch (recoverable: all 42 re-queued as pending, re-mines nightly or in one manual cloud research_lane run).

Decisions/state appended 2026-07-11 (DATA chat: SELF-HEAL layer built)

• A unified SELF-HEAL orchestrator is built (affirmology-studio/api/self_heal.py, commit 577829a, PUSH HELD until the mirror render window is clear). Jeff's ask: stop his time going to manually re-kicking builds after a failure or a new intake, get our systems robust enough to finish themselves. THE PROFILE STORE IS THE LEDGER (no new table): no-chart+birth = pending chart (free), chart+no- Understanding = pending Understanding (~$0.20), both = done. A failure just stays pending and retries next pass (boot or manual), self-healing + retry-safe across chats, code sessions, and Render restarts. Understanding heal is bounded per pass, skip-existing, STOPS on the first Anthropic credit error (never drains the pool), and defers while a manual understanding batch runs. Boot runs the unified pass; new intakes auto-sequence chart -> Understanding. Live ledger any chat can read: GET /api/self-heal/ status; manual trigger POST /api/self-heal/run. Env kill-switches UNDERSTANDING_AUTOHEAL(_MAX/_PACE). Verified (mocked): ledger bucketing, stop-on-credit, batch guard, credit detector all pass. Coverage is currently 37/37 Understandings (Jeff was right, first Understanding is the done-for-everyone default), so the first heal pass on deploy is a no-op; value is future intakes + failures.
• FUTURE (not built, documented next step): the "another WAY after failure" = local<->cloud render/ build fallback routing (hybrid node). Self-heal today covers "another TIME after failure" (retry next pass). The build_understanding cloud endpoints ALREADY EXIST (/api/build-understanding, -batch, /status) with cost discipline; self-heal is the autonomous layer on top of them.

Decisions/state appended 2026-07-11 (DATA chat: audio-suite NAMING + first-constellation shape)

• AUDIO NAMING SYSTEM decided (Jeff, 7/11). The mechanical/software frame (Subconscious OS, "update pending", version numbers, streaks) stays in the APP UI layer; the AUDIOS themselves get SACRED, rooted names (brand voice: light/water/root/breath/hearth; never "install/claim/sync/boost"). Names are a DISPLAY LABEL only (stable internal structure_id underneath), so they can be user-renamable, seasonally rotated, or A/B'd with ZERO engine risk. Old working labels (Daily Pair / Night Install / Morning Claim / S5 / S11) are RETIRED.
• TWO-REGISTER system (Jeff's refinement, layer by intensity, don't pick one): Register 1 = elemental/CALM for the everyday rhythm; Register 2 = heroic/CINEMATIC for the charged + deep. The name's register itself SIGNALS the experience intensity before play.
• First-constellation v1 mapping (adjustable):
• ☀ Morning = First Light (R1) · 🌙 Sleep = The Night Tide (R1) · 🧘 Meditation CATEGORY = The Stillness (R1) holding The Release (EFT/tapping) + The Seeing (visualization).
• 🔥 Movement/gym = The Forge (R2, forged through effort; "The Kindling" KILLED as too weird) · 🌀 Deep-journeys CATEGORY (R2) uses The Descent (deep dive, NOT nightly sleep), The Becoming (hero's-journey long), The Passage (Chiron, wound->gift); journey-suite = The Turning. More transformational structures to come; each gets its own R2 heroic name.
• The Rising = a NEW audio (freed from movement): a stronger, EVOLVING Origin Soul Song, the Sun/core-identity anthem. Distinct from the current locked origin_soul_song (the fixed, chart-pure entry gift). The Rising is the LIVING version, it deepens / re-renders as the person evolves (transits, weekly context, journey), so the identity anthem never goes stale. Ties to membership-longevity + the weekly-context input. Concept, not yet a built structure.
• Plus a new READING at entry (The Mirror = flagship first reading). Made-at-entry: Morning + Sleep + a Movement + the Meditation pair + a reading; more on discovery or greyed-out + rolled out occasionally with an invite (retention loop).
• WEEKLY CONTEXT PROMPT is a new generation input (engine build item). At entry each week the person picks a context (selection box) + optional type-or-voicenote. That context shapes THAT week's audios + reading alongside the chart/Understanding, so the suite is current and re-earnable, not chart-static. Slots into the Understanding's context_store; voicenote reuses the on-device transcription the Oracle mic already plans (Parakeet/sherpa-onnx + Whisper), NOT ElevenLabs. Refines Cowork's App Experience Master Plan (constellation system + Get Audios); keep the two aligned.

STANDING RULE (Jeff, 2026-07-06): ALL UNDERSTANDING GENERATION RUNS IN THE CLOUD

Jeff called it a FAIL that Understanding generation ran on his laptop (tied to him keeping the Mac awake while leaving). NEW HARD RULE: every future Understanding build (standard AND deep six-lens) must be generated CLOUD-SIDE, fire-and-forget, never dependent on a local machine. BUILD NEEDED: a cloud endpoint on the Studio (Render) e.g. POST /api/build-understanding {person, depth} that runs the Sonnet synthesis server-side and stores to profile.understanding (standard) or KB (deep, separate), plus a batch/cron to run it across the list. Until that endpoint exists, do NOT start laptop-bound Understanding batches. This is the priority infra fix.

Decisions/state appended 2026-07-11 (Cowork app-experience chat: App Experience Master Plan v1)

• Affirmology_AppExperience_MasterPlan_v1.md is the LEAD doc for the app experience (the UX layer on top of Map v2; the Wave 1 floor, Daily Pair + feedback taps, is unchanged and protected). Specs the locked 3-tab nav (Explore / Listen / Oracle), the Listen daily-suite home, Oracle kept pure chat, the full Explore IA (constellations, request desk, journeys, blueprint charts+reports, relationships, practices, voice lab, feedback hub, profile, help), Constellation v1 (Aries first, reward = guide activation), the learning backend, status messaging + overnight delivery loop, and Atlas Creator Lab notes. Monday v0 build table inside (9 items with owners).
• Cheap wins decided: playback speed = CLIENT-SIDE rate control saved per audio type (no re-render); pause width = per-person setting honored at NEXT render; NEW ENGINE ASK for DATA = /render-patch (re-voice one edited beat verbatim, splice at the beat boundary; our paced renders are already beat-concatenated, so this is feasible and makes belief/metaphor updates near-free).
• Corrections so nobody re-researches: in-app voice notes = on-device Parakeet via sherpa-onnx (server Whisper interim), NOT ElevenLabs; chart graphics already exist (pdf_chart_svg.py, all five systems + the synthesis diagram); relationship/Penta engine already built; written blueprint sections shape from the stored Understanding (pennies, once), never council re-runs.
• Explore ships Monday as a v0 portal: live rows = Constellations (Aries), Get Audios (minimal), Feedback Hub, Profile (thin), Help; everything else = named forming cards. Constellations REFLECT practice, never gate it (locked ethics).

STANDING RULE - cross-chat handoffs (Jeff, 2026-07-12). Applies to EVERY chat.

Jeff is the human router between several chats (Cowork/DATA, the app/code chats, script chats). When work moves between chats, the deliverable is a SELF-CONTAINED FENCED CODE BLOCK printed in the response, ready to copy and paste. Never hand Jeff only a file card or a viewer link and expect him to relay it. He cannot copy-paste from those, and it burns a full round trip every single time.

Every handoff block must carry: 1. a one-paragraph CONTEXT blurb, so the receiving chat needs zero prior history 2. explicit REFERENCES: the relevant .md filenames + the dated _CHAT_SYNC.md block 3. the actual request / decisions / spec 4. the SPLIT OF WORK: what is already built, and what the receiving chat must build

Write the file AND print the paste block. Both. Every time. The job is faster human/AI co-creation. Making Jeff re-type or re-summarize is the single biggest avoidable inefficiency in this project.

BACKLOG / PARKED - Sphere V3 audio visualizer (logged 2026-07-12, Jeff)

Rapid Flow Sphere V3 ($39 one-time, 14-day free trial): https://www.rapidflow.shop/products/sphere-v3 An audio-reactive visualizer that runs as a DAW plugin (VST3 / AU / AAX) on the master bus, with built-in video capture up to 4K, vertical + landscape exports, custom presets, and logo embedding.

• Cannot be used inside the Maya app. No SDK, no API, no headless/CLI. It only runs in a desktop DAW GUI, so it cannot render per-user Sacred Audio visuals (every audio is unique to one person) and the license is not for redistribution.
• Where it IS useful: a studio/production tool for ONE-TO-MANY pieces rendered once and shown to everyone: investor + mission videos, social clips, the Origin Soul Song hero piece, a few "featured" journey audios, the demo.
• Caution: ships with a photosensitive-epilepsy / strobing warning. Our use case is sleep, meditation, heart coherence. Any preset we adopt must be slow, calm, non-flashing.
• STATUS: PARKED. Jeff will buy the DAW + this toolkit when affordable and get it into someone's hands. Not blocking anything.
• The in-app answer is different and better: build a real-time on-device reactive visualizer (Web Audio AnalyserNode -> WebGL / Three.js shader, emerald+gold). It reacts to whatever is playing, so it works for every user's unique track at zero render cost, and it can be CHART-DRIVEN (colors from their placements, geometry from their HD type) which Sphere can never be. Prototype not yet built.

Decisions appended 2026-07-12 (DATA chat, Jeff testing build 32 live)

• NO DAILY EMAIL UNTIL AUG 13 (Jeff). The Wave 1 daily email is deferred until Aug 13 unless bandwidth frees up earlier. Nobody obsesses about email copy now; the spec sits in Affirmology_CoworkHandoff_NameFlags_Wave1Email_v1.md for when it activates.
• NEW RULE: never overwrite a wheel/app ART key on R2, always publish a NEW versioned key. The 7/12 figure cleanup overwrote wheel/figures/ in place; the Cloudflare CDN kept serving the OLD starred bytes (and the phone in gemini) to the app. Fix shipped: clean art republished under v2_ keys + alignment.json repointed (hot-reload, no build). Same lesson class as the corpus.db overwrite rule.
• WHEEL FEEL FEEDBACK (Jeff, on device): the current Wheel reads as a flat left/right list; no rotating-through-a-space-wheel feeling, and swipe-down-to-the-Sun does not work. Jeff: "a bigger image map needs to be created here." This is the next Wheel design/build iteration (rotation carousel + vertical Sun gesture + possibly a large pre-rendered sky map). Owner: app chat + Cowork design; log ideas before building.
• NAV BAR ICONS: Jeff finds the top nav icons ugly; HE is sourcing better examples for Cowork.

Decisions appended 2026-07-13 (Corpus briefing chat, mining request from app-experience chat)

• The "56 pending" somatic/attachment/IFS batch is ALREADY MINED. The nightly research cron cleared the entire backlog on the night of 7/12 to 7/13: 269/269 done, 395 craft entries, corpus at 25,570 records. No manual catch-up run is needed. Cancer (1/4/5), Libra 3, Scorpio 5/8 (Parts Council, Reclaim The Exiled), Gemini 6/8/11, and Pisces 5 are UNBLOCKED for template writing now. Ideal Parent Figure, Hakomi probes, IFS deep mechanics, trauma-informed audio design, and neuroception/felt-safety voice specs all landed with 0.82 to 0.92 confidence.
• All 8 mining lanes from the Zodiac Content Report queued (46 new backlog entries, in R2). Money psychology/Klontz (7, Taurus), performance + speaking (6, Leo), grief protocols incl. Weller, Worden, dual-process (7, Scorpio Grief Passage), dream science + TMR Paller/Rasch (6, Pisces), voice intimacy/ parasocial (5, retention lever), program design craft (6, all challenges), career/vocation (6, Capricorn), HD + GK delivery craft (3, Aquarius/Sagittarius). The nightly cron mines them; at the observed ~56/night rate, expect all lanes done within 1 to 2 nights.