Project_Velocity/.Agent Context/Sprint 1/Founder CRM and Platform Delivery Pack/14 - Reconciliation Matrix_ Sprint 1 Fact Table vs Current Repo.md

Reconciliation Matrix_ Sprint 1 Fact Table vs Current Repo

Date: 2026-04-18
Status: Draft
Owner: Sagnik
Reviewers: Sayan, Sourik
Scope: Reconcile the 2026-04-12 Sprint 1 Fact Table against the current Project_Velocity repo and live architectural truth
Purpose: Show what is current, stale, partial, or missing before new founder planning proceeds.
Decision Boundary: This matrix updates planning truth. It does not by itself deprecate older docs, but it does mark where they are no longer authoritative.

1. Status Key

Status	Meaning
exists and current	Present in repo and still aligned with current product truth
exists but stale doc	Repo reality has moved beyond what the old fact table claims
partially built	Significant implementation exists, but the target outcome is still incomplete
missing	No material implementation or planning truth exists yet

2. Executive Findings

• The biggest stale area is not infrastructure. It is documentation about CRM and route maturity.
• Oracle is much more real than older planning snapshots imply.
• CRM is more real than old placeholder assumptions implied, but still not canonical.
• Inventory has product surface presence, but not canonical data persistence.
• The current repo has overlapping lead and interaction models that require planned convergence.

3. Reconciliation Matrix

Area	Old Fact Table Claim	Current Repo Truth	Status	Founder Interpretation
FastAPI neural core	Real and done	backend/main.py is a real unified runtime	exists and current	Keep as current truth
CRM backend routes	Older fact table claimed missing/placeholder route state in places	backend/api/routes_crm.py is implemented and mounted	exists but stale doc	CRM exists, but its schema and surface are still incomplete
Oracle backend routes	Older fact table previously treated Oracle route files as empty	backend/api/routes_oracle.py exists and is mounted	exists but stale doc	Oracle is a real runtime, not a placeholder
Oracle Canvas	Not emphasized as production-grade reality	app/src/app/oracle/page.tsx is a significant real surface	exists but stale doc	Oracle is already a first-class product surface
CRM frontend truth	WebOS shell strong but CRM incomplete	still true	partially built	CRM has building blocks, but no top-level operational shell
Frontend module map	no top-level CRM described	app/src/App.tsx exposes dashboard, oracle, sentinel, inventory, catalyst, settings	exists and current	CRM is still missing as a top-level module
Lead data model	implied need for canonical structure	repo still contains leads, leads_intelligence, store mocks, and mapper layers	partially built	canonical convergence still missing
Interaction model	implied incomplete	repo has chat_logs, omnichannel_logs, consent, vault, CCTV and perception artifacts	partially built	needs unification under one client graph
Inventory DB	not described as complete	inventory UI exists, but canonical persistence does not	partially built	planning still required
Prompt inventory	called out as missing	still true	missing	founder monolith should absorb this later
Comfy/infra truth	infra substantially achieved	still true, but truth is split across multiple docs	partially built	needs one canonical founder-owned operator reference

4. Explicitly Resolved Overlaps

4.1 leads_intelligence vs leads

• current truth: both exist conceptually in repo/database planning
• interpretation: neither is final canonical truth
• planning decision: both become legacy feeders into future crm_* domains

4.2 omnichannel_logs vs chat_logs

• current truth: there are at least two interaction/log concepts in play
• interpretation: current repo expresses both operational chat history and broader communication logging
• planning decision: both map into the future intel_* interaction graph

4.3 mocked inventory UI vs onboarding assets

• current truth: inventory UI exists and is polished, but backed by mock/store truth
• current repo also contains real onboarding CSV/catalog assets
• planning decision: inventory remains partially built until canonical persistence is defined

4.4 Oracle runtime vs Oracle template DB planning

• current truth: Oracle runtime is real today
• current gap: Oracle JSON template/component corpus is still a separate planned deliverable
• planning decision: runtime and template DB are distinct concerns and should not be collapsed into one claim

5. What Is Actually Missing

• top-level CRM module and IA
• canonical client graph schema
• canonical import layer
• canonical inventory persistence model
• unified client 360 aggregation contract
• formal employee/user target model beyond current users_and_roles
• founder-owned monolithic truth document

6. What This Means For Founder Planning

The founder pack should not start from “nothing exists.” It should start from:

• strong runtime exists
• partial CRM exists
• Oracle exists
• infra exists
• planning coherence does not

7. Bottom Line

The old fact table was directionally useful, but it is no longer authoritative in several key areas. The current founder planning layer should treat the repo as materially advanced and focus on convergence, not reinvention.