alaivOS — Consolidation Audit Report¶
Generated: April 13, 2026 (end of Omega v2.7 session, after PC-reboot recovery)
Scope: Project directory C:/Users/Pp/Downloads/Flutter Code/ and its files/ subdirectory (~600+ .md docs accumulated over ~3 months).
Goal: Collapse the scattered KB into one authoritative canonical set so Omega (web chat) and builders (Antigravity/Claude Code) can share the same shared-brain (Cloudflare KV ALAIVOS_KB → key master-kb).
1. DELIVERABLES¶
17 flat canonicals + 2 meta-files, all in the project root (no subfolders, per instruction):
| # | File | Size | Role |
|---|---|---|---|
| 1 | index.md |
7.8 KB | Entry point, reading order, locked decisions digest |
| 2 | CLAUDE.md |
23.6 KB | Builder constitution, locked decisions, service mappings, P0 digest, coding standards |
| 3 | PRODUCT_SCOPE.md |
25.9 KB | Full module/feature inventory (14 modules, Checkup, AQ, Sports, multi-agent) |
| 4 | PRICING_AND_TIERS.md |
29.0 KB | 7 tiers, 120+ gates × 7, Ghost credits, Group Plans, Checkup cadence |
| 5 | TRIAL_AND_ONBOARDING.md |
22.0 KB | 21-day trial, Day 0/14/28 Checkups, Laiv Rules of Engagement |
| 6 | PROGRESSIVE_INTERVIEW_DESIGN_SPEC.md |
25.4 KB | 45 questions × 11 traits, mandatory, Elite unlock |
| 7 | ANTI_ABUSE_SPEC.md |
23.2 KB | 7 defense layers, device fingerprint, phone verify Day 14 |
| 8 | GHOST_PROTOCOL.md |
31.8 KB | Gemma 4 E4B brain, Kokoro TTS, Checkup relay, credit model |
| 9 | VOICE_AND_AI.md |
23.0 KB | Model stack, AMI, Kokoro-first pipeline, multi-agent, PromptAssembler |
| 10 | DATA_PIPELINE.md |
23.0 KB | 3 servers, 290 cities, 19 TomTom keys, DDG/Photon/Nominatim, 5-layer traffic composite, sports cache |
| 11 | MAP_MODULE_SPEC.md |
20.6 KB | 5 views, glass POI pins, overlays (rain/AQ/wind/compass), 5-layer search |
| 12 | INFRASTRUCTURE.md |
18.8 KB | Hetzner/Bishop/Cloudflare/Supabase/Anthropic/Twilio/Firebase/RevenueCat/Stripe |
| 13 | LAUNCH_PLAYBOOK.md |
12.1 KB | Critical path, J-tasks, builder sprints, feature-gap status, 45-day timeline |
| 14 | LEGAL_AND_PRIVACY.md |
32.7 KB | Privacy, ToS, health disclaimers, law enforcement, canary, DMCA, third-party AI |
| 15 | WEBSITE_SPEC.md |
25.1 KB | alaivos.com, hub, 13 pages, design system, pre-launch Kappa TODOs |
| 16 | MARKETING_AND_GROWTH.md |
30.5 KB | 23 psychographic clusters, channels, World Cup plan, referrals, badges |
| 17 | LESSONS_LEARNED.md |
22.6 KB | Battle-tested P0 lessons with symptom/root-cause/fix/where |
| — | FILE_INDEX.md |
35.4 KB | Full catalog of every .md file in the project + status tag |
| — | FILE_AUDIT_REPORT.md |
this file | This report |
Total canonical surface: ~417 KB authoritative markdown. Replaces ~180+ scattered docs as the single source of truth per topic.
2. METHODOLOGY¶
Parallel synthesis by specialized subagents, one per target canonical. Each agent was briefed with:
- TIER 0 (newest wins) —
OMEGA_V2_7_SESSION_HANDOVER.md(today) +MASTER_PENDING_ITEMS.md(today's Omega compilation) - TIER 1 —
OMEGA_V2_6,V2_5,V2_4_FINALhandovers (recent locked decisions) - TIER 2 — Primary canonical from
files/(the previous consolidation for that topic) - TIER 3 — Current sprints + patches/addenda specific to that topic
- TIER 4 — Historical detail sources (absorb non-superseded content)
Supersession rule: Newer decisions delete older ones; older content is preserved only if never superseded. No merging of old+new versions of the same information — pick the newest.
Output: Each agent wrote its target file directly with Write, returning a ≤300-word audit digest that I've consolidated below.
3. PER-CANONICAL AUDIT¶
3.1 CLAUDE.md — Builder constitution¶
Absorbed: files/CLAUDE.md (Mar 31, Omega v2.4); OMEGA_V2_5/V2_6/V2_7 handovers; MASTER_PENDING_ITEMS; alaivos-builder-rules.md; TEST_GUIDANCE.md; SONNET_HANDOFF_CORRECTED.md; HANDOVER_next_architect.md; OMEGA_AND_PROJECT_INSTRUCTIONS.md; how_to_feed_claude_agents.md; KB_DOCUMENT_UPDATES_V2_4.md; KB_UPDATE_PACKAGE.md.
Key changes vs. prior: 3 new BUILT v1.0 features (Laiv Checkup, AQ stack, 31-league Sports); Observer 10→11 rules; 3 AES-256-GCM encrypted tables; 6 onboarding fields → Supabase; workflow spine (Cloudflare KV, business track split); metrics refreshed (~970+ tests, 120+ gates, 45 days to launch).
Contradictions resolved: Qwen 2.5 → 3.5; "9B on-device" → Ghost-only via Gemma 4 E4B; AMI always-resident → dynamic one-at-a-time; Bishop GPU server → mini PC; AiProvider 3-valued → 2-valued; gate count 115 → 120+; ElevenLabs TTS → Kokoro-first; "LAUNCH-READY" certification → 0 hard blockers + 11/14 crit / 5/5 partial / 3/3 call done.
3.2 PRODUCT_SCOPE.md¶
Absorbed: V4_FINAL (primary); V2_TAP_AUDIT; V3_LIFECYCLE_AUDIT; earlier PRODUCT_SCOPE + alaivOS_Complete_Scope + alaivOS_project_scope_v1 + alaivOS_Feature_Roadmap_Phases_ABC + mode_system_architecture + capsule_schemas_spec + travel/ereader/sports_fan/quiniela/health_household_habits/marketplace/laiv_system_prompt specs; V2_5/V2_6/V2_7 handovers. Superseded: 5 prior PRODUCT_SCOPE variants + alaivOS_Complete_Scope + project_scope_v1 + Feature_Roadmap. Contradictions resolved: gate count 107→119; tests 3,500→~4,333 (3,543 baseline + TAW); CX43 kept (not downgraded to CX23); Bishop mini PC (not GPU server); Sessions now on dock (closes V4 gap D6); Cloud Gemini dead.
3.3 PRICING_AND_TIERS.md¶
Absorbed: V2.7/V2.6/V2.5/V2.4 handovers; primary files/PRICING_AND_TIERS.md (Mar 24); DELTA_CORE_ANNUAL_PRICE_CHECK; sprint4_addendum_pricing; banking_pricing_table; tier_feature_matrix_with_trials; alaivOS_RevenueCat_Config; alaivOS_Tier_Differentiation_Matrix; alaivOS_tier_feature_matrix; alaivOS_Feature_Tier_Map; alaivOS_AI_Tier_Strategy; alaivOS_pricing_tiers_definitive; alaivOS_Pricing_Strategy_v3_Final + _v3_Refinements; alaivos_financial_model_summary; competitive_analysis_actionable; tiered_ai_capability_spec; feature1_revised_draft_messaging.
Key numbers locked: Starter Free / Spark $3.99 / Core $7.99 / Pro $14.99 / Elite $23.99. Annual = pay 10 get 12. Core annual ≈ $6.65/mo (not $6.66). Banking add-on $0.99/mo/bank (not $2.99 or $1.49). Elite includes 50D/20I Ghost credits + 1 bank. Group Plans 50% off cap 6, open to anyone. Checkup cadence Elite 1mo / Pro 2mo / Core 3mo / Spark 6mo / Starter none. 120+ gates.
Superseded: all prior PRICING_AND_TIERS editions + v2/v3 strategy docs + Feb 25 Core $4.99 / Premium $24.99 / 9-module tier + old RevenueCat core/pro/premium SKUs.
Contradictions resolved: banking price → $0.99; "Premium" renamed "Elite"; TTS menu Standard/HD/Custom (Clear/Natural dead); Checkup is tier-baked not credit-gated.
3.4 TRIAL_AND_ONBOARDING.md¶
Absorbed: V2.7/V2.6 handovers; primary files/TRIAL_AND_ONBOARDING.md; ADDENDUM_TRIAL_STRATEGY_2026; ADDENDUM_PROGRESSIVE_INTERVIEW_DESIGN_SPEC; MANDATORY_INTERVIEW_PATCH; SPRINT_ALPHA_REINSTALL_PERSISTENCE; SPRINT_DELTA_ONBOARDING_AUDIT; DELTA_QUICK_CHECK_PROGRESSIVE_INTERVIEW.
Locked: 14 Pro + 7 Elite = 21 days guaranteed (up to 28 with referrals). 45-question mandatory interview. Day 14 mandatory phone verify (SHA-256 hash + Twilio). 8 onboarding fields → Supabase user_profiles. Day 0/14/28 Checkup flow (Day 28 FREE). TAW10 username catch-block fix.
Superseded: trial_model_locked, trial_strategy_2026, trial_enchantment_blueprint, trial_experience_design, trial_touchpoint_strategy, trial_touchpoints_detailed, sprint10_BETA_addendum_onboarding, sprint_G2/G3/G4/B3/B4/B8, builder_prompt_phaseA6, beta_onboarding_persona_legal, MANDATORY_INTERVIEW_PATCH, ADDENDUM_PROGRESSIVE_INTERVIEW.
Contradictions resolved: 25 Qs → 45; "up to 28 days" → 21 guaranteed; old G-3 7-screen questionnaire → DEAD CODE; Elite unlock = 6/11 traits OR Day 14 fallback; Ghost trial access still not available except Days 5-7 surgical moment; referral days bank against Elite phase.
Residual risk: Layer 4 (Source D fixed-day questions) flagged UNVERIFIED per V2_7 — Delta must verify before launch.
3.5 PROGRESSIVE_INTERVIEW_DESIGN_SPEC.md¶
Absorbed: primary + ADDENDUM + MANDATORY_INTERVIEW_PATCH + DELTA_QUICK_CHECK + SPRINT_ALPHA_PROGRESSIVE_INTERVIEW + CLAUDE.md v2.7; earlier profiling sources (user_profiling_spec, alaivOS_dimensional_profiling_draft, sprint_G8_cluster_profiling, Gemini research prompts).
Locked: 45 Qs (24 A + 14 B + dynamic C + 7 D). 11 traits: Life Stage, Financial Attitude, Health Relationship, Sports Relationship, Learning Style, Work Rhythm, Planning Style, Communication Style, Tech Comfort, Aesthetic Preference, Privacy Posture. Elite unlocks at 6/11 populated (required: Life Stage + Financial + Health + 3 of remaining 8) OR Day 14 deterministic fallback.
Unverified claims flagged (§0 in file): Layer 4 engine presence; 45-Q bank built; card integration in Prism/Wind Down; Source D scheduler firing; markQuestionnaireCompleted() wiring; 31 tests passing; 5 ARB i18n keys × 21 locales; old G-3 interview_screen.dart confirmed dead. Reported built by Gamma March 26 but NOT re-audited after TAW/v2.6/v2.7.
3.6 ANTI_ABUSE_SPEC.md¶
Absorbed: primary files/ANTI_ABUSE_SPEC.md v1.1 (March 30); project-root CLAUDE.md v2.7.
7 layers locked: (1) device fingerprint silent signup, (2) phone verify mandatory Day 14 SHA-256 + server salt, (3) progressive interview mandatory GT #10, (4) no data export Trial/Starter (Spark+ gate), (5) investment score Day 21, (6) behavioral detection analytics-only, (7) RLS + real collision at INSERT (TAW10 fix).
Also absorbed: Twilio + RevenueCat backstop; trial_devices / phone_verifications / trial_signals schemas; 8 user_profiles fields; Starter nudge throttling (72h, 3-strike mute); law-enforcement confirm-match-only row; privacy.html + terms.html + Play/App Store disclosures; 15 locked decisions; 23 ARB keys; 4 defense-chain walkthroughs.
3.7 GHOST_PROTOCOL.md¶
Absorbed: V2_7 handover; MASTER_PENDING_ITEMS; GHOST_PROTOCOL_UNIFIED; SPRINT_EPSILON_CHECKUP_RELAY / _KOKORO_EVAL / _COTURN_CX43 / _ENCRYPT_PLAINTEXT; SPRINT_ALPHA_CHECKUP_PIPELINE / _E2EE_GHOST / _GHOST_UI / _MODEL_FIX_GHOST_CREDITS; EPSILON_SESSION_CANONICAL_NOTES; EPSILON_GHOST_AND_AIRPORT; Ghost_Protocol_App_Spec + Website_Spec + Visual_Identity_Spec; ghost_agent_beta_spec. Locked: Gemma 4 E4B Q4_K_M @ 12 tok/s on CX43, 10 GB RAM, native function-calling EN/ES/PT, audio, vision. Qwen 3.5 9B → fallback. Kokoro 82M (StyleTTS 2, Apache 2.0, CPU) as Ghost TTS. Credits-only gate for in-app Ghost skills. Checkup = tier-baked (Spark 2/yr, Core 4/yr, Pro 6/yr, Elite 12/yr). Checkup pipeline: device strip → Gemma 4 E4B digest → Anthropic Batch (Sonnet 4.6), ~$0.012/checkup, return_token opaque, 72h auto-delete, Day 28 FREE. Thinking mode OFF by default on Gemma Instant. CX43 stays €17/mo. Superseded: GHOST_PROTOCOL_UNIFIED, ghost_protocol_v2_1_LOCKED, ghost_protocol_v3_0_spec, sprint_B12_ghost_protocol, epsilon_jtask_update, plus all "Qwen 9B" / "ElevenLabs Ghost TTS" claims. Contradictions resolved: Ghost model Qwen 9B → Gemma 4 E4B + 9B fallback; Ghost voice ElevenLabs/WaveNet → Kokoro 82M; Checkup credit-gated → tier-baked; AiProvider → {local, ghost}.
3.8 VOICE_AND_AI.md¶
Absorbed: V1_1_AI_ROADMAP (primary); V1_1_ROADMAP; V2_7/V2_6/V2_5 handovers; SPRINT_EPSILON_KOKORO_EVAL; SPRINT_ALPHA_PIPER_TTS; SPRINT_ALPHA_CHECKUP_PIPELINE / _SKILL_AUDIT; SPRINT_S1_SKILL_ROUTER; SPRINT_S2_OBSERVER_AGENT; SPRINT_S3_PLANNER_EXECUTOR; SPRINT_S4_PERSONALITY_UI; SPRINT_BETA2_ASK_LAIV; SPRINT_BETA_TTS_DEAD_CODE_SEVERANCE; SPRINT_B13_sovereign_voice_asr; sprint_A5/A6/A2; sprint_B10_gguf_runtime; SPRINT_GAMMA_LOCAL_MODEL_IMPL; tiered_ai_capability_spec; alaivOS_qwen3_tts_strategy; alaivOS_ai_model_assessment; qwen_mistral_evaluation; gguf_runtime_research_report; laiv_system_prompt_spec; alaivOS_AI_Prompt_Library_v2; alaivOS_phone_ai_server_hub_architecture. Locked: On-device = Qwen 3.5 only (0.8B/2B/4B practical; E2B/E4B future). AMI dynamic load/unload, one at a time. CDN tier-based filenames + manifest.json v3. Smart Flutter Responses (10 handlers). Kokoro-first voice pipeline (EL ref → Kokoro style vector → Piper on-device + Voxtral 3B Ghost HD). Multi-agent v1.0: SkillRouter (17 skills multilingual after TAW), Observer (11 rules), Planner (6 action types), Executor (8 module actions, confirmation-gated, rollback), Personality (5 presets + 4 sliders). Brain Distillation stays v1.1 (~3,500 examples, ~$475-710, 4-6 weeks on Bishop).
3.9 DATA_PIPELINE.md¶
Absorbed: PROPRIETARY_DATA_PIPELINE (primary); V2_7/V2_6/V2_5 handovers; all EPSILON handovers + v2.0/v1.1; all SPRINT_EPSILON_ (PIPELINE, PIPELINE_FULL, TRAFFIC_DEPLOYMENT, PHOTON, ENRICHMENT_ENGINE, SPORTS_CACHE v1/v2, 3SERVER_WEATHER, LOAD_TOMTOM_KEYS, THREE_TASKS, BONSAI_EVAL, CDN_RENAME, SETUP, SERVER_W2, NEXT); SPRINT_ALPHA_AQ_CHECKUP_TRAVEL; SPRINT_BETA2_AQ_ALERTS / _TRAFFIC_ENGINE; SPRINT_POI_MEGA_HARVEST; GEMINI_SPORTS_API_RESEARCH; GEMINI_AIRPORT_INFRASTRUCTURE; traffic_intelligence_deployment_plan; traffic_deployment_guide; epsilon_traffic_deployment; city_tiers; city_data_prioritization; motorcycle_traffic_calibration; airport_transit_data_pipeline; holiday_school_calendars; map_routing_traffic_strategy_LOCKED + predecessors.
Locked:* 3 servers (ghost-01 CX43 €17/mo, cx23 €4/mo, cx23-b €4/mo). 19 TomTom keys, ~47,500 daily API budget. 290 Full + 1,178 expansion queue Apr 29. 3 tiers only (Standard/Full/Maintenance — Seed/Lite dead). Weather + AQ via Open-Meteo every snapshot (free, ≤30 min stale). RainViewer live-net only for radar. Overpass backbone + DDG nightly 22:00 UTC ~43,200 q/day. Photon Cloudflare Worker proxy (self-host failed, 17 GB > 38 GB disk). Nominatim for geocode, UA alaivOS/1.0. ZERO paid API deps (Google/Foursquare/Yelp cancelled; Google credit $300 USD dart-define kill-switch expires Jun 5). 5-layer traffic composite: Catmull-Rom Gold / linear Standard, chips in MINUTES not %, calendar 20 countries, events proximity+decay+attendance, largest-only no stacking, 8 files, 58 tests. Sports cache 31 leagues ghost-01:8300 1hr TTL. Checkup Relay ghost-01:8100 Anthropic Batch.
3.10 MAP_MODULE_SPEC.md¶
Absorbed: primary files/MAP_MODULE_SPEC.md; V2_7/V2_6/V2_5; SPRINT_CRA trio; W1-W4 map waves; SPRINT_ALPHA_MAP_POI_PERSONA / _MAP_SEARCH; SPRINT_BETA1_MAP_OVERLAYS + OVERLAY_FIXES; SPRINT_CONTACT_MAP_PINS; SPRINT_R3_ALPHA_OFFLINE_MAPS / _BETA_GEOFENCING; SPRINT_TILE_PREFETCH; SPRINT_NAV1-4; SPRINT_BETA2_TRAFFIC_ENGINE.
Gates locked: interactiveMap Starter (map+voice nav+motorcycle+POI+My Places+Trip+offline maps FREE); mapTrafficPatterns Spark; mapLiveTraffic Core; mapTrafficNavigate Pro; mapFamilySharing Spark (PARTIAL: local-only); contactMapPins Spark.
Contradictions resolved: pin-zoom V2_5 progressive (dots z10-12, emoji z13-14, cards z15+) → V2_6 glass-pins-only (dots z13-15, mini z16, detail z17+) due to clutter; factor chips minutes not %; Catmull-Rom cubic Gold / linear Standard; Cloud Gemini dead; family sharing PARTIAL; Foursquare/Yelp cancelled; TomTom live-only not primary; Qwen 3.5 + Gemma 4 E4B per v2.7.
3.11 INFRASTRUCTURE.md¶
Absorbed: LAUNCH_STATUS_AND_TASKS; EPSILON v1.1 + v2.0 handovers; EPSILON_SESSION_CANONICAL_NOTES + SETUP; SPRINT_EPSILON_COTURN_CX43 / _CHECKUP_RELAY / _ENCRYPT_PLAINTEXT / _SERVER_W2 / _NEXT / _2B_CDN_UPLOAD; EPSILON_CDN_RENAME; ZETA_CX22_PROVISIONING; SPRINT_WAVE2_INFRA_GAPS; CLOUDFLARE_WORKER_BANKING / _TASKS; SPRINT_ALPHA_FCM_PUSH; SUPABASE_SPRINT16_MIGRATION; SPRINT_ALPHA_CALL_SIGNALING; SPRINT_BETA1_CALL_SESSION; alaivOS_phone_ai_server_hub_architecture; alaivOS_web_architecture_complete. Locked: Ghost-01 services table (Ollama 11434, ghost-router internal, sports-cache 8300, checkup-relay 8100, nginx 443, coturn 3478/5349). Bishop = mini PC Ryzen AI 9 HX 370, 64 GB DDR5, no GPU, no CUDA. CX43 kept. Scaling >500 subs → second node; >1000 → GEX44 GPU €184/mo. Monthly: ~€28 + $3. Contradictions resolved: Bishop GPU server → mini PC; Qwen 9B on-device → Gemma 4 E4B Ghost-only + 9B fallback; AiProvider → {local, ghost} only.
3.12 LAUNCH_PLAYBOOK.md¶
Absorbed: 27 source docs — MASTER_PENDING_ITEMS; V2.7/2.6/2.5/2.4 handovers; alaivOS_Launch_Playbook; LAUNCH_STATUS_AND_TASKS; EXHAUSTIVE_TASK_LIST_V4; master_task_list; j_tasks_updated / _launch_prep / _automation_script; external_services_j_action_list; sonnet_j_tasks_briefing + _full_briefing + _deputy_briefing; launch_roadmap_to_worldcup; alaivOS_prelaunch_checklist + _master_implementation_checklist; PENDING_ITEMS_AUDIT + _COMPLETE; STORE_METADATA_DRAFT; THREE_LAYER_LAUNCH_AUDIT; FIELD_TEST_TASK_LIST_V1; exhaustive_task_list_by_role + _v3_status; worldcup_2026_strategy. Locked: 45 days to launch. Zero hard blockers. 11/14 crit + 5/5 partial + 3/3 call done. 12 J-tasks unblocked + 5 blocked-on-external + 2 pre-revenue urgent. Canary due April 19. Contradictions resolved: Delta v2.7 "no hard blockers" overrides earlier "v1.0 LAUNCH-READY" claim; call sprints (v2.6 "status unknown") confirmed DONE; metrics ~970+ tests / 31 leagues / 290 AQ cities / 120+ gates; infra cost ~€28 + $3.
3.13 LEGAL_AND_PRIVACY.md¶
Absorbed: LEGAL_AND_WEBSITE; alaivOS_privacy_policy + _terms_of_service + _health_disclaimer_framework; V2_7/V2_6/V2_5; omega_legal_review; sprint_G1_legal_settings; ANTI_ABUSE_SPEC disclosure requirements; gemini_research_prompt_legal_framework. Locked: Citerius Holdings LLC (Wyoming) legal entity. Anthropic API as org Citerius Holdings LLC. Universal E2EE via libsignal_protocol_dart. Health data NEVER cloud, NEVER in Ghost prompts. Encrypted-at-rest tables: shared_gift_notes, web_ai_content, ai_command_queue (AES-256-GCM). 8 onboarding fields persisted to Supabase (disclose). Warrant canary re-sign due April 19. DMCA designated agent pending via Registered Agents Inc. Pre-launch updates required: privacy.html (phone hash collection + third-party AI processing line for Anthropic Checkup); terms.html (21-day trial, Day 14 phone verify mandatory, mandatory interview); lawenforcement.html (phone hash row, confirm-match only). Attorney review before significant user volume. Cross-border tax counsel before first dollar.
3.14 WEBSITE_SPEC.md¶
Absorbed: KAPPA_FULL_WEBSITE_AUDIT + _PATCH_ONEPASS + _WEBSITE_HANDOVER; SPRINT_KAPPA_WEBSITE; alaivOS_web_architecture_complete; alaivOS_website_update_complete + _marketing_guide + _marketing_update_v2 + _Website_Update_Guide; WORLDCUP_LANDING_PAGE_SPEC; Waitlist_Google_Sheets_Setup; website_restructure_instructions + _about_page_draft + _contact_page_draft + _builder_pricing_update_guide; builder_prompt_website_corrections; Ghost_Protocol_Website_Spec + _Visual_Identity_Spec.
Pages covered (13): index.html, index-prelaunch.html, about, contact, privacy, terms, health, billing-terms → /billing, lawenforcement, canary, pgp, /worldcup (pending), + portals hub.alaivos.com + my.alaivos.com.
Pre-launch updates flagged (K1–K9): privacy.html (phone + third-party AI), terms.html (trial conditions), lawenforcement.html (phone hash row), /worldcup build, 7 legal-page deploy, launch-day swap, hub.alaivos.com deploy, outdated "Ultra 9B on-device" + missing voice-nav copy corrections, canary re-sign.
Contradictions resolved: "Ultra 9B on-device" dead; "offline AI" → "on-device AI"; Streams → Chat + Notes; voice nav + motorcycle + map = FREE all tiers (was missing from pricing copy); Ghost = credit model (per-skill dead); copyright © 2026 alaivOS (never Citerius in copyright line).
3.15 MARKETING_AND_GROWTH.md¶
Absorbed: primary files/MARKETING_AND_GROWTH.md; MASTER_PENDING_ITEMS; V2_7; alaivOS_store_listings; alaivOS_Badge_System_Spec; worldcup_2026_strategy + _venue_schedule; alaivOS_Reddit_Posts; competitive_analysis_actionable; alaivOS_psychographic_cluster_map; alaivOS_lifestyle_marketing_plan; alaivOS_Content_Marketing_Strategy; alaivOS_Marketing_Guidance; alaivOS_Marketing_Audit_v1; alaivOS_Launch_Playbook; launch_roadmap_to_worldcup; sonnet_marketing_pricing_pivot_brief; competitive + deep_research + gemini_research prompts.
Locked: 23 psychographic clusters preserved (Tier 1: Juggler, Expecting, Scrapper, Optimizer, Chef, Student · Tier 2: Guardian, Elder, Hustler, Tracker, Healer, Professional, Voyager · Tier 3: Faithful, Creative, Newlywed, Snowbird, Steward, Backbone, Clerk, Queen, Warrior, Sovereign) + 3 GDL WC sub-clusters. 8 channels. 12-week World Cup beat plan. K-factor >1 Quiniela target. Household viral loop. Group Plans 50% off cap 6 + referral tiers. Affiliate 50%/20%. 5-wave schedule + Sovereign parallel. 4-badge system. 45-day launch roadmap. Post-launch waitlist-to-badge swap. 10-item anti-positioning copy QA.
3.16 LESSONS_LEARNED.md¶
Absorbed: CLAUDE.md P0 section; STALE_CONTENT_AUDIT; P0_blank_screen_diagnosis; P0_clean_slate_rebuild; P0v2_deep_diagnostic_from_masterplan; P0v2_multi_device_failure; loose_bolts_audit; alpha_ diagnostic (size_and_gemma_theory, clean_slate_rebuild, final_p0_plus_gemma_removal, gemma_db_sprint15, navigation_fix, p0_final_fix, revised_diagnostic, updated_overlay_test); architect_response_to_alpha_report; download_failure_report; seeding_audit; beta_trace_report_request; masterplan_review_shortcomings; masterplan_v3_review; beta_masterplan_fix; beta_circadian_background_fix; hotfix_download_wakelock_chunked + _import_screen_blocker + _onboarding_keyboard; HOTFIX_BETA1_SHARED_EVENTS_RENAME + _TABLE_NAME. Format:* Each lesson = Symptom / Root cause / Fix / Where in code / Discovered-in. Grouped by category: Widget/State, Performance, Data/SQLite, Auth/Onboarding, Maps/POI, Networking, Localization, Privacy/Branding, Models/AI, Lint/GT.
3.17 index.md¶
Role: Entry point. Distills locked decisions (12 bullets), reading order (per role: Omega / Builder / Epsilon / Delta / Kappa), canonical-files table (16 rows), workflow & shared brain (Cloudflare KV ALAIVOS_KB), out-of-scope, provenance.
4. CROSS-CUTTING SUPERSESSION CHAINS¶
| Domain | Before this consolidation | After (single source of truth) |
|---|---|---|
| Pricing | 8+ files (strategy v2/v3/v3_refine/v3_final, tier matrices, RevenueCat config, banking price variants, pricing_tiers_definitive) | PRICING_AND_TIERS.md |
| Ghost | 6+ files (UNIFIED, v2_1_LOCKED, v3_0, B12 sprint, App/Website/Visual Identity specs, agent_beta_spec) | GHOST_PROTOCOL.md |
| Trial | 7+ files (trial_model_locked, _strategy_2026, _enchantment_blueprint, _experience_design, _touchpoint, _touchpoints_detailed, ADDENDUM_TRIAL_STRATEGY_2026, MANDATORY_INTERVIEW_PATCH) | TRIAL_AND_ONBOARDING.md |
| AI/Voice | 5+ files (V1_1_AI_ROADMAP, V1_1_ROADMAP, tiered_ai_capability_spec, alaivOS_qwen3_tts_strategy, alaivOS_ai_model_assessment, qwen_mistral_evaluation, gguf_runtime_research_report) | VOICE_AND_AI.md |
| Pipeline | 10+ files (PROPRIETARY_DATA_PIPELINE + 8 traffic/city/airport/holiday specs + EPSILON handovers) | DATA_PIPELINE.md |
| Map | 5+ files (MAP_MODULE_SPEC, map_routing_traffic_strategy_LOCKED, alaivOS_map_route_strategy, _map_routing_traffic_strategy, + 20+ sprint files) | MAP_MODULE_SPEC.md |
| Legal | 5+ files (LEGAL_AND_WEBSITE, alaivOS_privacy_policy, _terms_of_service, _health_disclaimer_framework, omega_legal_review) | LEGAL_AND_PRIVACY.md |
| Website | 7+ files (KAPPA audits x3, alaivOS_web_architecture_complete + _website_update_complete + _marketing_guide + _marketing_update_v2 + _Website_Update_Guide + WORLDCUP_LANDING_PAGE_SPEC) | WEBSITE_SPEC.md |
| Launch tasks | 13+ files (EXHAUSTIVE_TASK_LIST_V4, master_task_list, j_tasks_ x4, external_services_j_action_list, sonnet_j_tasks_ x3, sonnet_deputy_briefing, launch_roadmap_to_worldcup, alaivOS_prelaunch_checklist + _master_implementation_checklist, PENDING_ITEMS x2, STORE_METADATA_DRAFT, THREE_LAYER_LAUNCH_AUDIT, FIELD_TEST_TASK_LIST_V1) | LAUNCH_PLAYBOOK.md |
| Marketing | 10+ files (MARKETING_AND_GROWTH primary + 9 cluster/lifestyle/content/audit/reddit/WC/badges/pivot/roadmap docs) | MARKETING_AND_GROWTH.md |
5. RESIDUAL RISKS (Delta must verify before launch)¶
Flagged by V2.7 and preserved in the canonicals:
| Claim | Source | Canonical flagging | What to verify |
|---|---|---|---|
| 17/17 Laiv skills execute real writes | V2_7 §10 | PRODUCT_SCOPE.md, VOICE_AND_AI.md |
Say "I spent $45 on groceries" → check Money module row |
| Progressive Interview Layer 4 (Source D fixed-day) built | V2_7 §10 | PROGRESSIVE_INTERVIEW_DESIGN_SPEC.md §0 |
Check interview questions actually fire Days 1-14 |
| Onboarding data read by Laiv | V2_7 §10 | TRIAL_AND_ONBOARDING.md |
Grep for reads of birthday/sex/roadblock/management_style |
| All ~200 empty catches resolved (TAW8) | V2_7 §10 | LESSONS_LEARNED.md |
Grep for catch(e) {} + silent failures |
| Wave 1/2 fix sprints execution status | MASTER_PENDING_ITEMS §"Wave 1 Fix Sprints" | LAUNCH_PLAYBOOK.md |
J confirmation needed (SPRINT_ALPHA_FCM_PUSH, SPRINT_BETA1_E2EE_CHAT, SPRINT_BETA2_EVENTS_MONEY, SPRINT_GAMMA_QUICK_FIXES, SPRINT_WAVE2_INFRA_GAPS) |
| Family calendar sharing | V2_7 Delta audit | PRODUCT_SCOPE.md, MAP_MODULE_SPEC.md |
Marked local-only PARTIAL; no cloud sync service yet |
| iOS DND activation | V2_7 Delta audit | PRODUCT_SCOPE.md |
Platform limitation; in-app only |
| Bank connection | V2_7 Delta audit | PRODUCT_SCOPE.md, PRICING_AND_TIERS.md |
Deferred — blocked on Plaid/Belvo keys |
6. STRAY DUPLICATE FILES (flagged by FILE_INDEX agent, cleanup recommended)¶
files/OMEGA_V2_4_SESSION_HANDOVER_FINAL (1).md— duplicate offiles/OMEGA_V2_4_SESSION_HANDOVER_FINAL.md. Delete the(1)copy.files/SPRINT_TG_BETA (1).md— duplicate offiles/SPRINT_TG_BETA.md. Delete the(1)copy.
7. NOTES ON PROCESS¶
- PC reboot mid-run: An unexpected PC reboot interrupted Wave 3 + Wave 4 dispatch. Seven agents had already completed and flushed to disk atomically (Write is all-or-nothing); they survived intact. Five were still in flight; I verified on return which 12 canonicals existed and re-dispatched the missing 5 (MAP_MODULE_SPEC, ANTI_ABUSE_SPEC, PROGRESSIVE_INTERVIEW_DESIGN_SPEC, MARKETING_AND_GROWTH, FILE_INDEX). All 17 + FILE_INDEX + index.md are now complete.
- TIER 0 path discrepancy: Early agents were pointed at
C:/Users/Pp/Downloads/OMEGA_V2_7_SESSION_HANDOVER.mdandC:/Users/Pp/Downloads/MASTER_PENDING_ITEMS.md(where the user originally dropped them via@). Some agents found them atC:/Users/Pp/Downloads/Flutter Code/files/OMEGA_V2_7_SESSION_HANDOVER.mdandC:/Users/Pp/Downloads/Flutter Code/MASTER_PENDING_ITEMS.mdinstead. Content was identical; no data loss. - FILE_INDEX snapshot timing: The FILE_INDEX agent scanned before MARKETING_AND_GROWTH.md and index.md landed, so it reports 15/17 canonicals present. This is a timing artifact only — all 17 are now on disk and FILE_INDEX can be refreshed on next touch.
- No .html/.docx/.pdf read: Per user instruction, only
.mdfiles were consulted. HTML legal pages, PDF research reports, and DOCX exports remain infiles/for reference but weren't absorbed directly; their content was already distilled into derived .md files which WERE absorbed.
8. NEXT STEPS¶
- Push
master-kbto Cloudflare KV. Per V2.7 workflow spine, Epsilon should concatenate or index all 17 canonicals and push to theALAIVOS_KBnamespace under keymaster-kb. Additional keys per V2.7:current-sprint,sprint-results,sprint-log,delta-findings,arch-decisions. - Enable Cloudflare MCP in Omega web chat. So Omega can read/write KV without paste.
- Delete stray duplicates (§6 above).
- Delta verification sprint for the 8 residual risks in §5 before launch.
- Pre-launch Kappa TODOs per
WEBSITE_SPEC.md§0 +LEGAL_AND_PRIVACY.md: privacy.html phone + third-party AI line, terms.html trial conditions, lawenforcement.html phone-hash row, /worldcup landing, 7 legal-page deploy, hub.alaivos.com deploy, canary re-sign (April 19). - Business track kickoff (per V2.7 §9): separate directory
alaivos-biz-core/+ separate KValaivos-biz-kb+ separate Claude.ai project.
9. VERIFICATION¶
All 17 canonicals atomically written; each ends with a proper sign-off (Supersedes list / Maintained-by-Omega note / End-of-file marker). File sizes in §1. Full 45-question bank preserved in PROGRESSIVE_INTERVIEW_DESIGN_SPEC.md. Full feature-gate matrix preserved in PRICING_AND_TIERS.md. Full trait model preserved. Full Ghost-01 service map preserved. Full 3-server pipeline topology preserved. Full 23-cluster marketing spec preserved. All privacy policy language absorbed (source files still on disk for verbatim reference per LEGAL_AND_PRIVACY.md footer).
Zero content loss detected. The 17-file canonical set is architect-ready for distillation and unification with the web KB.
Generated by consolidation audit at end of Omega v2.7 session. See index.md for canonical entry point. See FILE_INDEX.md for full historical-file catalog with status tags.