Documentation Inventory — Knowledge Hub

Archived S504 (ruling R2): superseded by Pagefind search + generated stats — see the docs-site keep-docs-in-sync skill. S205C-era snapshot retained below.

Documentation Inventory — Knowledge Hub

⚠️ HISTORICAL SNAPSHOT (noted S491, 22/07/2026). This inventories the pre-migration in-repo docs/ tree of the canonical repo. Most of that tree has since moved into this private docs-site (or been retired); the canonical repo’s docs/ now holds only reference/testing/. Counts and paths here are S205C-era (28/04/2026) and earlier — do not trust them against the filesystem. Archive candidate.

Date: 28 April 2026 (S205C — §2.1 refreshed, §2.5b runbooks added, §1 totals re-counted; non-tracked-doc tables left at S153 snapshot) Session: 117 original → S152B WP1/WP2 partial refresh → S153 full rewrite of §2.1–§2.5 → S205C <-/kpf:refresh-reference-docs> seed run Trust level: Authoritative for §1 totals (re-counted S205C), §2.1 (docs/reference/), and §2.5b (docs/runbooks/). §2.2–§2.5 (specs, audits, design, operations) and §2.6–§2.16 are S153/older snapshots — counts may drift from the filesystem; refer to docs/generated/codebase-stats.md for live numbers and run /kpf:refresh-reference-docs to refresh tracked-doc headers. Purpose: Comprehensive catalogue of every documentation file in the project, with purpose, audience, currency, and recommended actions.

---

Changelog

• S318 (06/06/2026): New ontology register entry docs/ontology/33-reference-item.md — Layer-5 reference_item entity for the ID-75 reference layer (WP-F, BI-22), authored in the 32-q-a-pair.md shape. Companion register edits in the same change: 13-ingest-source.md (§6.3/OQ-T1 re-homing of rss_feed/url_import to reference_items.ingestion_source + reserved adopted_from_reference), 29-cocoindex-source-kind.md (remote_url baseline value), and docs/ontology/README.md (new “Where does new data live?” decision rule with the Q1a/Q1b split + register-table rows for 33). The docs/ontology/ register (33 CV/entity files + README) is not yet enumerated as a §2 directory section in this inventory — flagged for the next full refresh.
• S205C (28/04/2026): First end-to-end run of /kpf:refresh-reference-docs. All 11 tracked canonical docs (§2.1 highlighted rows + §2.5b runbooks) now carry <!-- Last verified: S205C (28/04/2026) against commit 3a563758 --> headers (replacing the S205B WP2a seed placeholder). §2.1 table extended with a Last verified column and a new entry for auth-hooks.md (24 total). §2.5b added to inventory the three tracked runbooks plus the new ci.md introduced in kh-prod-readiness-S7. §1 Summary file counts refreshed: docs/reference/ 23 → 24, docs/runbooks/ 0 → 4, docs/operations/ 11 → 20 (operations table refresh remains out of scope for this run).
• S153 (07/04/2026): Full rewrite of §2.1 (docs/reference/), §2.2 (docs/specs/), §2.3 (docs/audits/), §2.4 (docs/design/), and §2.5 (docs/operations/) against filesystem ground truth. Matches the post-S152B-WP1 layout (23 reference files, 12 specs, 44 audits grouped into 5 sub-folders, 4 design files, 9 operations files). §1 Summary totals refreshed. §2.6–§2.16 not touched — they remain pre-S117 snapshots.
• S152B WP1 (07/04/2026): Executed the S151 reference-folder audit. 22 files moved out of docs/reference/ — 14 to docs/audits/{sector-intelligence,ai-eval,classification,strategic,content}/, 3 to docs/operations/ (including taxonomy-change-runbook.md), and 5 to .planning/.archive/.reference/ or .planning/.archive/.research/. New file docs/reference/prompt-improvement-methodology.md created from the archived prompt-improvement docs. New file docs/reference/state-of-the-product-history.md created by S152B WP10 split. Final docs/reference/ count: 23 Live reference files. The §2.1 table below still reflects the pre-move state — full rewrite is tracked as follow-up WP in the next session’s handoff.
• S152B WP2 (07/04/2026): Partial cross-reference sweep. Stale refs to the moved files updated in docs/reference/state-of-the-product.md, docs/reference/product-roadmap.md, docs/reference/entity-type-taxonomy-spec.md, docs/reference/product-differentiation-audit.md, docs/specs/si-prompt-refinement-skill-spec.md, and docs/specs/ai-eval-phase2-classification-skill-spec.md. The inventory itself and state-of-the-product-history.md (frozen snapshot) are intentionally NOT fully rewritten.

1. Summary

Major Changes Since Last Inventory (S108)

The project underwent a large-scale archive reorganisation between S108 and S117. The key structural changes are:

1. .planning/specs/ removed entirely — all 37 specs moved to .planning/.archive/.specs/
2. .planning/feasibility/ removed — 18 files moved to .planning/.archive/.feasibility/
3. .planning/research/ removed — 79 files moved to .planning/.archive/.research/
4. .planning/audits/ removed — files moved to .planning/.archive/.audits/
5. .planning/session-extracts/ removed — 20 files moved to .planning/.archive/.session-extracts/
6. .planning/continuation-prompts/ removed — files moved to .planning/.archive/.continuation-prompts/
7. .planning/ root post-mvp-backlog.md removed — moved to .planning/.archive/.specs/
8. docs/audits/ reduced from 18 to 6 — s105 audit files and wp4 verification moved to .planning/.archive/.audits/
9. docs/session-exports/ removed entirely — s77 and s91 exports moved to .planning/.archive/.session-exports/
10. docs/continuation-prompts/ reduced from 98 to 14 — older prompts (S1-S103) moved to .planning/.archive/.continuation-prompts/
11. docs/specs/ reduced from 67 to 9 — completed/historical specs archived to .planning/.archive/.specs/
12. docs/reference/ gained 3 new files — ai-visibility-policy.md, product-backlog.md (renamed from critique-product-backlog.md in S150), product-backlog-completed.md (renamed from critique-product-backlog-completed.md in S150); lost 1 (critique-gap-crossref.md archived)
13. docs/design/ lost 1 file — design-brainstorm-session.md archived to .planning/.archive/.specs/
14. Archive path changed — from .planning/archive/ to .planning/.archive/ (dot-prefixed)

Total File Counts

Location	Files	Notes
docs/reference/	24	Live reference (S205C re-count; +1 since S153 — auth-hooks.md, see §2.1)
docs/runbooks/	5	Tracked runbooks for prod-readiness + sidecar ops (see §2.5b)
docs/specs/	12	Active implementation specs (S153 snapshot — re-count pending, see §2.2)
docs/audits/	44	25 top-level + 19 in 6 topic sub-folders (S153 snapshot, see §2.3)
docs/design/	4	Design system (includes 1 PDF, see §2.4)
docs/operations/	20	Production + operational runbooks (S205C re-count; +9 since S153, §2.5)
docs/continuation-prompts/	14	S153 snapshot — older prompts archived to .planning/.archive/
docs/generated/	4	Auto-generated stats (2 JSON + 2 Markdown)
docs/product-functionality/	4	Skeleton — README + 3 templates
docs/database/	2	Schema snapshot + Q&A backup
docs/uat/	5	UAT scenario results and research
docs/client-documentation/	22	Client-facing docs (9 .docx, 1 .pdf, 12 .md)
.planning/codebase/	7	Codebase analysis documents
.planning/reviews/	52	Codebase review runs and findings
.planning/.archive/	442+	Archived documents (10 subdirectories; recent S132–S196 add ~58 files)
.claude/checks/	10	Quality check files
Project root	1	CLAUDE.md
Total active	189	Excluding archived files (S205C re-count)
Total archived	500+	In .planning/.archive/ (≥58 added S205C archive sweep)
Grand total	689+	All documentation files

Trust Level Distribution

Trust Level	Count	Percentage
Authoritative	~40	~24% of active
Historic	~127	~76% of active
Archived	~442	Separate category

Currency Distribution (Active Files Only)

Currency	Count	Percentage
Current	~35	~21%
Needs update	~10	~6%
Outdated	~5	~3%
Historic	~117	~70%

---

1.5 Refresh tooling — /kpf:refresh-reference-docs

A reusable slash-command (/kpf:refresh-reference-docs) refreshes the 11 canonical reference docs tracked in lib/docs/tracked-reference-docs.ts against the live codebase, schema, env vars, and MCP inventory. Each refresh also regenerates this inventory file so its tables match the on-disk state.

When to run:

• The /update-docs Step 6.5 drift list crosses ~5 entries.
• Quarterly, even if no drift signal has fired.
• After a major schema change, env-var rotation, or AI-stack overhaul.

What it does:

• Spawns 4 parallel kpf-reference-doc-refresher agents (Schema / AI–classification / Data flow / Runbooks). Each agent diffs claimed facts in its partition against the live codebase, surgically rewrites drifted sections, and bumps the doc’s <!-- Last verified: ... --> header in the same edit (the WP2b edit-coupled freshness guard rejects body-only edits).
• Orchestrator regenerates docs/reference/documentation-inventory.md (this file) as a final non-agent step.
• Single review-ready commit; nothing is auto-pushed.
• Typical runtime: ~30–60 minutes.

Underpinning artefacts:

• ~/.claude/commands/kpf/refresh-reference-docs.md — slash-command definition (orchestrator, partition table, inline workflow).
• ~/.claude/agents/kpf-reference-doc-refresher.md — agent definition.
• lib/docs/tracked-reference-docs.ts — single source of truth for the 11-doc tracked list. Adding or removing a tracked doc requires updating both this constant and the partition table inside the slash-command file.

Upstream drift detector: /update-docs Step 6.5 — emits the drift list that triggers this command.

---

2. Inventory by Directory

2.1 docs/reference/ — Core Reference Documents (24 files)

Authoritative, long-lived reference. Every file in this directory is intentionally kept live — historical/archived content lives under .planning/.archive/.reference/ instead. The 8 reference docs marked tracked in the Last verified column are also enforced by the WP2b edit-coupled freshness guard (__tests__/docs/reference-doc-edit-coupled-freshness.test.ts) and listed in lib/docs/tracked-reference-docs.ts. Post-S152B WP1 audit layout, S205C seed-run refresh.

File	Purpose	Audience	Trust	Last verified	Feeds Into
SCHEMA-QUICK-REFERENCE.md	Retired — redirect stub. Schema truth now lives in supabase/types/database.types.ts + database-overrides.ts (ID-47.9)	agent	deprecated	retired — ID-47.9	All areas
state-of-the-product.md	Current tech stack, architecture, feature state (canonical — 798 lines post-WP10 split)	both	authoritative	current (S153)	All areas
state-of-the-product-history.md	Append-only history of session additions S93–S151 (WP10 — frozen reference)	both	historic	frozen	All areas
product-roadmap.md	Active roadmap — pre-launch blockers, forward-looking only	both	authoritative	current (S153)	All areas
product-backlog.md	Main product backlog — items not currently on the roadmap (renamed from critique-product-backlog.md S150)	both	authoritative	current (S150)	All areas
product-backlog-completed.md	Completed items archive extracted from product backlog	both	historic	current	All areas
documentation-inventory.md	This file — comprehensive catalogue of all project documentation	both	authoritative	current (S205C)	All areas
_archive/task-list-discipline.md	Per-field write discipline for the workflow ledgers — ID-34. Archived S491 (low-read/stale; discipline now enforced by ledger-cli soft-warn budgets + update-ledgers skill)	agent	historic	archived (S491)	All areas
decision-graph.md	Intended-vs-current rosetta stone — phase-0 decisions (S227–S233) mapped to ratified+built state (S234–S251). Promoted + rebuilt S267 from phase-0-investigation/0.9-decision-graph.md (now a stub)	both	authoritative	current (S267)	All areas
deferral-register.md	Forward register of every [DEFERRED-v1.1]/[DEFERRED-v2]/[DEFERRED-POST-LAUNCH] item (source + rationale) — seeds the new product roadmap. Sibling of decision-graph.md (S267)	both	authoritative	current (S267)	All areas
ai-integration-strategy.md	Master AI strategy — four-layer model	both	authoritative	tracked — S205C	ai-integration
ai-integration-layers.md	Four-layer AI architecture reference guide	both	authoritative	tracked — S205C	ai-integration
ai-visibility-policy.md	AI visibility and branding policy — invisible infrastructure	both	authoritative	current	All areas
auth-hooks.md	Auth hooks reference — getAuthorisedClient() discriminated union, authFailureResponse() HTTP routing	agent	authoritative	current	administration
classification-prompt.md	Classification prompt — loaded by Python pipeline at runtime	agent	authoritative	tracked — S205C	knowledge-organisation
classification-architecture.md	Classification architecture — Pass 1/Pass 2 flow, diagnostic questions	agent	authoritative	tracked — S205C	knowledge-organisation
entity-type-taxonomy-spec.md	Entity type taxonomy — source of truth for entity classification	agent	authoritative	tracked — S205C	knowledge-organisation
field-consumer-dependency-map.md	Field → consumer dependency map for schema changes	agent	authoritative	tracked — S205C	All areas
data-entry-points.md	Canonical reference for every write path into content_items + Appendix D change_reason vocabulary	agent	authoritative	tracked — S205C	All areas
two-pass-validation-architecture.md	Two-pass classification validation architecture	agent	authoritative	current	knowledge-organisation
two-pass-validation-results.md	S151 WP3 measurement results — precision 41.7% against ≥70% launch target	both	authoritative	current (S151)	knowledge-organisation
sector-intelligence-pathway.md	Sector Intelligence product pathway	both	authoritative	current	sector-intelligence
prompt-improvement-methodology.md	Prompt-improvement methodology (created S152B WP1 from archived prompt-improvement docs)	agent	authoritative	current (S152B)	ai-integration
product-differentiation-audit.md	Sales-ready competitive analysis and demo readiness	human	authoritative	current	bid-management
mcp-app-build-guide.md	Technical guide for building, testing, deploying MCP Apps	agent	authoritative	current (S151)	ai-integration
ux-principles.md	The 13 governing UX principles, extracted from ADS v1.0 §6	both	authoritative	current	All areas

Changes from the S108/S117 inventory:

• Added (S152B WP1): state-of-the-product-history.md (split from canonical), prompt-improvement-methodology.md (new), data-entry-points.md (from elsewhere), classification-architecture.md, entity-type-taxonomy-spec.md, field-consumer-dependency-map.md, two-pass-validation-architecture.md, two-pass-validation-results.md, sector-intelligence-pathway.md.
• Removed from docs/reference/ (S152B WP1 — moved, not deleted): content-readiness-tracker.md → docs/operations/; taxonomy-change-runbook.md → docs/operations/ (was in reference briefly); si-feed-onboarding-strategy.md → docs/operations/; content-enrichment-candidates-s98.md → docs/audits/content/; strategic-product-assessment.md + strategic-assessment-gap-analysis.md → docs/audits/strategic/; sector-intelligence-phase0-research.md + sector-intelligence-phase1-research.md + si-extraction-audit-s149.md + si-gap-analysis-s149.md + si-gap-analysis.md → docs/audits/sector-intelligence/; ai-eval-gap-analysis-s149.md + ai-eval-research-s149.md + ai-touchpoint-map-and-improvement-strategy.md → docs/audits/ai-eval/; classification-framework-analysis.md + classification-investigation-report.md + entity-taxonomy-gap-analysis.md → docs/audits/classification/; content-readiness-review.md + 2026-03-03-knowledge-platform-design.md + prompt-skill-engineering-vision.md + systematic-prompt-improvement-approach.md → .planning/.archive/.reference/; context-graph-research.md → .planning/.archive/.research/.

2.2 docs/specs/ — Implementation Specs (12 files)

Active and recently-landed specs. Completed and historical specs are archived to .planning/.archive/.specs/ as part of the S108/S123 archive reorganisation — specs in this directory are expected to be referenced from the current roadmap or from an in-flight WP.

File	Purpose	Trust	Currency	Feeds Into
demo-bootstrap-spec.md	Pre-launch demo environment bootstrap (S136, roadmap §1.3) — seed.sql + bootstrap-instance.sh + onboard-client.sh	authoritative	current (S136)	administration
si-prompt-refinement-skill-spec.md	Prompt refinement skill (§2.2 PRE-LAUNCH BLOCKER post-S151 elevation) — entity precision recovery	authoritative	current (S151)	knowledge-organisation
ai-eval-phase2-classification-skill-spec.md	AI evaluation Phase 2 — classification skill evaluation layer	authoritative	current (S144)	ai-integration
mcp-evaluation-spec.md	MCP evaluation Layer 5 spec (L1/L3/L4 live — see scripts/mcp-eval/)	authoritative	current	ai-integration
bid-testing-strategy-spec.md	Bid workflow UAT scenario specs	authoritative	current	bid-management
template-driven-completeness-spec.md	Template-driven completeness — phases complete + remaining	authoritative	current	bid-management
workspaces-launcher-spec.md	Workspaces launcher design spec	authoritative	current	content-management
workspaces-launcher-gap-analysis.md	Gap analysis for the workspaces launcher spec against current implementation	historic	current	content-management
silent-failure-prevention-spec.md	Silent failure prevention — sb() wrapper + ESLint rules (S151/S152C)	authoritative	current (S152C)	administration
silent-failure-migration-s152c.md	S152C migration plan for the silent failure prevention rollout	historic	current (S152C)	administration
structured-logging-spec.md	Structured logging spec — domain.entity.action naming, logBestEffortWarn helper	authoritative	current (S152C)	administration
python-pipeline-config-env-vars-spec.md	Python pipeline configuration and environment variable reference	authoritative	current	knowledge-organisation

Changes from S108 (67 → 12): bid-session-coverage-spec.md, bid-session-implementation-plan-s116.md, jsonb-normalisation-spec.md, jsonb-data-model-assessment-s116.md, and upload-page-review-spec.md archived after their work completed. New specs landed via S136 (demo-bootstrap-spec.md), S144 (ai-eval-phase2-classification-skill-spec.md), S149 (si-prompt-refinement-skill-spec.md), S151 + S152C (silent-failure-prevention-spec.md, silent-failure-migration-s152c.md, structured-logging-spec.md), and earlier sessions (python-pipeline-config-env-vars-spec.md, workspaces-launcher-*). pipeline-parity-spec.md archived after S151 (now at .planning/.archive/.specs/pipeline-parity-spec.md).

2.3 docs/audits/ — Audits, Reviews, and Research (44 files)

Grew from 6 files (S108 inventory snapshot) to 44 files. The growth is driven by (a) S151 comprehensive codebase audit wave producing ~22 audit documents, (b) S152A/B production-bug investigations, and (c) the S152B WP1 reference-folder reorganisation which moved all time-bounded audit/research material out of docs/reference/ into topic subfolders under docs/audits/.

Top level — session-scoped audit and investigation documents:

File	Purpose	Session	Status
s151-ai-integration-audit.md	AI integration end-to-end audit	S151	historic
s151-api-silent-failure-audit.md	API silent-failure patterns audit — feeds into silent-failure-prevention-spec.md	S151	historic
s151-clean-routes-reaudit.md	Clean-routes re-audit after first-pass fixes	S151	historic
s151-continuation-prompt-audit.md	Continuation-prompt archival recommendations	S151	actioned (S152B WP13)
s151-decision-responses.md	Liam’s Q&A responses to S151 findings (actioned S152B WP12)	S151	actioned
s151-context-graph-research-implementation-status.md	Context-graph research vs implementation status	S151	historic
s151-e2e-coverage-gaps.md	E2E test coverage gap analysis	S151	historic
s151-fail-fast-partial-response-decisions.md	Fail-fast / partial response decisions record	S151	historic
s151-guard-tests-adversarial-review.md	Adversarial review of S151 guard tests	S151	historic
s151-lint-cleanup-report.md	Lint-cleanup report for S151	S151	historic
s151-lint-triage.md	Lint warning triage for S151	S151	historic
s151-mcp-app-build-guide-gap-analysis.md	MCP app build guide gap analysis	S151	actioned (S152B P-1)
s151-phase5-verification.md	Phase 5 verification log	S151	historic
s151-prompt-skill-engineering-vision-coverage.md	Prompt skill engineering vision coverage check	S151	historic
s151-reference-folder-audit.md	Plan for the reference-folder reorganisation (executed S152B WP1)	S151	actioned (S152B WP1)
s151-silent-failure-recheck.md	Silent failure re-check after initial fixes	S151	historic
s151-silent-failure-root-cause-analysis.md	Root-cause analysis for S151 silent failure findings	S151	historic
s151-specs-gap-analysis.md	Specs gap analysis	S151	historic
s151-wave1-adversarial-review.md	Wave 1 adversarial review	S151	historic
s151-wave2-adversarial-review.md	Wave 2 adversarial review	S151	historic
s151-wp14-fk-indexes-and-perf.md	S151 WP14 FK indexes + performance investigation	S151	historic
s152a-bid-drafting-production-bugs.md	Three production bugs (#16 CharacterCount, #17 sync guard, #18 reload)	S152A	actioned (S152B WP14)
s152a-feed-sources-rootcause.md	Feed sources root-cause analysis	S152A	historic
s152a-wp2-phase4-verification-log.md	Phase 4 §8.0.8 verification log	S152A	actioned (S152B WP14)
s152b-wp15-test-infra.md	WP15 test-infrastructure audit (written S153; verification runs tracked in §6)	S152B / S153	partial — awaiting 3× —workers=3 verification

docs/audits/ai-eval/ — AI evaluation research (3 files, moved S152B WP1):

File	Purpose
ai-eval-gap-analysis-s149.md	AI eval gap analysis — source of truth for AI eval roadmap items (§3 in roadmap)
ai-eval-research-s149.md	AI eval research document
ai-touchpoint-map-and-improvement-strategy.md	Map of AI touchpoints + improvement strategy

docs/audits/classification/ — Classification audits (3 files, moved S152B WP1):

File	Purpose
classification-framework-analysis.md	Classification framework analysis
classification-investigation-report.md	Classification investigation report
entity-taxonomy-gap-analysis.md	Entity taxonomy gap analysis

docs/audits/content/ — Content audits (1 file, moved S152B WP1):

File	Purpose
content-enrichment-candidates-s98.md	Analysis of 68 sales_brief layer items for enrichment potential

docs/audits/design/ — Design critique sessions (6 files, moved main 184f505):

File	Purpose	Persona
design-critique-session-1.md	Dashboard, Login, Settings	Sarah
design-critique-session-2.md	Browse, Item Detail	James
design-critique-session-3.md	Review Queue, Coverage	Rachel
design-critique-session-4.md	Search, Q&A Library	Tom
design-critique-session-5.md	Create Content, Upload	all
design-critique-session-6.md	Bid Detail, Session, Guide	all

docs/audits/sector-intelligence/ — SI audits/research (5 files, moved S152B WP1):

File	Purpose
sector-intelligence-phase0-research.md	Phase 0 research for the Sector Intelligence pathway
sector-intelligence-phase1-research.md	Phase 1 research
si-extraction-audit-s149.md	SI extraction audit
si-gap-analysis-s149.md	Source of truth for remaining SI roadmap items (roadmap §2 references this)
si-gap-analysis.md	Earlier SI gap analysis

docs/audits/strategic/ — Strategic assessment (2 files, moved S152B WP1):

File	Purpose
strategic-product-assessment.md	Pre-client-access product positioning assessment
strategic-assessment-gap-analysis.md	Gap analysis against strategic assessment recommendations

2.4 docs/design/ — Design System (4 files)

Warm Meridian design system — unchanged since S108 in file count.

File	Purpose	Audience	Trust	Currency	Feeds Into
warm-meridian-philosophy.md	Design philosophy — foundational principles	both	authoritative	current	All areas
warm-meridian-implementation-spec.md	Implementation spec — semantic tokens, colour system, components	agent	authoritative	current	All areas
warm-meridian-identity.pdf	Visual identity reference	human	authoritative	current	All areas
warm-meridian-audit-report.md	Post-migration audit — zero raw Tailwind violations found	agent	historic	frozen	All areas

2.5 docs/operations/ — Operational Guides (11 files)

Grew from 6 → 9 files in S152B WP1 by relocating operational runbooks out of docs/reference/ where they had accumulated. S156 added database-rebuild-runbook.md. S168 WP1c added re-ingestion-quality-protocol.md.

File	Purpose	Audience	Trust	Currency	Feeds Into
production-setup-guide.md	Step-by-step Vercel + Supabase deployment guide	human	authoritative	current	administration
bid-library-import-guide.md	Guide for importing Q&A pairs from .docx files	both	authoritative	current	content-management
search-evaluation-guide.md	How to run and interpret the 24-case search evaluation suite	both	authoritative	current	search
e2e-test-flows.md	Comprehensive E2E test flow specifications	both	authoritative	current	administration
e2e-test-setup.md	E2E test data setup runbook	both	authoritative	current	administration
test-bid-resources.md	Test bid templates, portals, and sample documents	both	historic	frozen	bid-management
taxonomy-change-runbook.md	Runbook for taxonomy changes — 7-step chain with 2 manual steps (moved from reference S152B WP1)	agent	authoritative	current	knowledge-organisation
content-readiness-tracker.md	Central task tracker for pre-client-access quality gates (moved from reference S152B WP1)	both	historic	frozen	content-management
si-feed-onboarding-strategy.md	Tier 1/Tier 2 SI feed onboarding strategy (moved from reference S152B WP1)	both	authoritative	current	sector-intelligence
database-rebuild-runbook.md	End-to-end runbook for rebuilding a Supabase project (pilot/demo/client)	both	authoritative	current	administration
re-ingestion-quality-protocol.md	Six-step quality gate for re-ingestion: snapshot, compare, divergence assessment, DOCX checklist	both	authoritative	current	content-management

Note: docs/operations/ count drifted from 11 (S153) to 20 (S205C re-count). The §2.5 table above still reflects the S153 layout and is missing 9 newer files. A full §2.5 refresh is out of scope for the S205C seed run (operational runbooks are not in TRACKED_REFERENCE_DOCS); flag for next inventory wave.

2.5b docs/runbooks/ — Tracked Operational Runbooks (7 files)

Tracked runbooks introduced post-S153 by the kh-prod-readiness track. Three are in TRACKED_REFERENCE_DOCS (enforced by the WP2b edit-coupled freshness guard); ci.md is informational and not yet tracked.

File	Purpose	Audience	Trust	Last verified	Feeds Into
local-development.md	Local dev setup — .env.local single source, prod/staging opt-in via --env=prod, sandbox bypass for CLI commands	both	authoritative	tracked — S205C	administration
staging-refresh.md	Persistent staging branch refresh procedure (kh-prod-readiness-S3) + S8 user_profiles parity verification	agent	authoritative	tracked — S205C	administration
github-environments.md	GitHub Environments setup — case-sensitive Production/Staging, env-var allocation post WP-S5.1/S5.2	human	authoritative	tracked — S205C	administration
ci.md	CI workflow runbook — added kh-prod-readiness-S7 WP-G4.1 (PR-blocking ci.yml chain). Not in TRACKED_REFERENCE_DOCS.	both	authoritative	not tracked	administration
cocoindex-deploy.md	RETIRED (S298) — cocoindex sidecar Cloud Run deploy (historical only; Cloud Run decommissioned, pipeline now on-prem via Coolify — see onprem-b1-deploy.md). Documents the two boot-required DBs (COCOINDEX_DB_DSN secretKeyRef + COCOINDEX_DB LMDB path), in-memory volume choice + Filestore persistence OQ, deploy + verify-GREEN commands (ID-49.9).	human	historic	tracked — S271	administration
_archive/staging-coolify-cocoindex.md	Staging Coolify stand-up runbook — staging mirror of the on-prem cocoindex stack (ID-66.21). Archived S491 (stand-up executed; live staging-stack ops + idle-mode boot triage now in onprem-b1-deploy.md; stale-era project refs + pullmd content)	human	historic	archived (S491)	administration
secret-rotation.md	Pipeline + app secret-rotation runbook — inventory of rotatable secrets + their surfaces, the {127.18} dual-accept pattern, cadence + ownership, and the S424 leaked-staging-set first-rotation checklist (4a–4h). Names + surfaces only, no values. Recovered from the S454 exec-127-18 transcript, landed S458.	human	authoritative	landed S458	administration

2.6 docs/continuation-prompts/ — Session Handoffs (14 files)

Reduced from 98 files (S108) to 14. Older prompts (S1-S103) moved to .planning/.archive/.continuation-prompts/ (73 files).

Audience: agent | Trust level: historic | Currency: historic | Action: keep

File	Session
continuation-prompt-kh-104-cert-cards-diff-engine.md	S104
continuation-prompt-kh-105-frontend-audit-layer-fix.md	S105
continuation-prompt-kh-106-diff-actions-product-docs.md	S106
continuation-prompt-kh-108-product-docs-auto-gen.md	S108
continuation-prompt-kh-109-critique-implementation.md	S109
continuation-prompt-kh-110-dashboard-workspaces-reader.md	S110
continuation-prompt-kh-111-dashboard-phase3-diff-review.md	S111
continuation-prompt-kh-112-content-discovery-and-backlog.md	S112
continuation-prompt-kh-113-content-discovery-p2-p3-and-quick-wins.md	S113
continuation-prompt-kh-s114.md	S114
continuation-prompt-kh-s115.md	S115
continuation-prompt-kh-s116.md	S116
continuation-prompt-kh-s117.md	S117
continuation-prompt-template.md	Template

2.7 docs/generated/ — Auto-Generated Documents (4 files)

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
codebase-stats.md	Auto-generated codebase statistics (Markdown)	both	authoritative	current (S116)	keep	All areas
codebase-stats.json	Machine-readable codebase statistics	agent	authoritative	current (S116)	keep	All areas
mcp-inventory.md	Auto-generated MCP tool/resource/prompt inventory (Markdown)	both	authoritative	current (S116)	keep	ai-integration
mcp-inventory.json	Machine-readable MCP inventory	agent	authoritative	current (S116)	keep	ai-integration

2.8 docs/product-functionality/ — Product Documentation (4 root files + subdirectory docs)

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
README.md	Directory structure, templates, authoring guidance	both	authoritative	current (S108)	keep	All areas
_templates/user-journeys.template.md	Template for user journey documentation	agent	authoritative	current (S108)	keep	All areas
_templates/workflows.template.md	Template for system workflow documentation	agent	authoritative	current (S108)	keep	All areas
_templates/technical.template.md	Template for API/component technical documentation	agent	authoritative	current (S108)	keep	All areas

bid-management/ (3 files, last verified S188):

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
technical.md	API routes, components, hooks, DB tables, MCP tools	both	authoritative	current (S188)	keep	bid-management
user-journeys.md	Bid creation, drafting, and export user journeys	both	authoritative	current (S188)	keep	bid-management
workflows.md	Bid lifecycle state machine, 3-pass AI drafting pipeline	both	authoritative	current (S188)	keep	bid-management

content-management/ (3 files, last verified S188):

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
technical.md	API routes, components, DB tables, dedup, supersession, quality gate	both	authoritative	current (S188)	keep	content-management
user-journeys.md	7 user journeys with role permissions	both	authoritative	current (S188)	keep	content-management
workflows.md	9 data-flow workflows with state transitions	both	authoritative	current (S188)	keep	content-management

knowledge-organisation/ (3 files, last verified S188):

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
technical.md	Taxonomy, entities, layers, guides — components, hooks, API routes	both	authoritative	current (S188)	keep	knowledge-organisation
user-journeys.md	Taxonomy admin, entity management, guide authoring user journeys	both	authoritative	current (S188)	keep	knowledge-organisation
workflows.md	Taxonomy sync pipeline, entity lifecycle, layer validation, guide publishing	both	authoritative	current (S188)	keep	knowledge-organisation

Subdirectories with .gitkeep only: search/, quality-governance/, ai-integration/, administration/

Note: Several of the above subdirectories now contain real .md files beyond .gitkeep and need inventory entries added. This was identified during S188 content-management refresh but is out of scope for this update.

2.9 docs/database/ — Database Documents (2 files)

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
schema-2025-03-05.sql	Historical schema snapshot (context only, not executable)	agent	historic	outdated	keep	administration
qa-backup-pre-reingestion-20260319.json	Backup of Q&A pairs before S97 re-ingestion	agent	historic	historic	keep	content-management

2.10 docs/uat/ — UAT Results (5 files)

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
uat-scenario-1-results.md	UAT Scenario 1: Standard Selection Questionnaire results	both	historic	historic (S77)	keep	bid-management
uat-scenario-2a-results.md	UAT Scenario 2a results	both	historic	historic	keep	bid-management
uat-scenario-2b-results.md	UAT Scenario 2b results	both	historic	historic	keep	bid-management
s77-uat-research-synthesis.md	Consolidated findings from UAT Scenario 1 and 8 research agents	both	historic	historic (S77)	keep	bid-management
e2e-results-session-57.md	E2E test results from Session 57 (12 flows, all passed)	agent	historic	historic (S57)	keep	administration

2.11 docs/client-documentation/ — Client-Facing Documents (22 files)

Authored Markdown (3 files)

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
Knowledge Hub — Platform Overview.md	Client-facing platform overview document	human	authoritative	current (S191)	keep	All areas
Knowledge Hub — Claude Integration Guide.md	Client-facing Claude integration guide	human	authoritative	current (S107)	keep	ai-integration
sector-intelligence-analysis.md	Analysis of sector intelligence brief from client	both	historic	historic (S77)	keep	content-management

Client .docx Files (9 files)

Source bid library documents from client (Phew Design). Not individually inventoried — these are input data files, not documentation.

• 4 “2026 Audit” template .docx files (FAQs, Functionality, Implementation, Security)
• 3 “DRAFT 2026 Phew” template .docx files
• 1 Product KB Dev Brief .docx
• 1 Sector Intelligence Brief .docx

Markdown Conversions (8 files in markdown/ subdirectory)

Markdown versions of the .docx files above, produced by pandoc conversion for pipeline processing. All historic.

Other (1 file)

File	Purpose	Audience	Trust	Currency	Action
Telehouse-South-Fact-Sheet-2025.pdf	Client reference document (data centre fact sheet)	human	historic	historic	keep

2.12 .planning/codebase/ — Codebase Analysis (7 files)

All produced by Session 96 codebase review. Analysis date: 2026-03-20.

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
ARCHITECTURE.md	Architecture patterns, layers, data flow	agent	authoritative	current (S96)	keep	All areas
CONCERNS.md	Technical debt, known issues, risk areas	agent	authoritative	needs-update	update	administration
CONVENTIONS.md	Coding conventions, naming patterns, file structure	agent	authoritative	current (S96)	keep	All areas
INTEGRATIONS.md	External API integrations (Anthropic, OpenAI, Supabase)	agent	authoritative	current (S96)	keep	ai-integration
STACK.md	Technology stack details (languages, frameworks, versions)	agent	authoritative	current (S96)	keep	All areas
STRUCTURE.md	Full directory layout with file-level detail	agent	authoritative	needs-update	update	All areas
TESTING.md	Testing patterns, frameworks, mock strategies	agent	authoritative	needs-update	update	administration

2.13 .planning/reviews/ — Codebase Reviews (52 files)

Structured codebase review runs from Session 84/91. All historic.

Audience: agent | Trust: historic | Currency: historic | Action: keep

Category	Count	Contents
2026-03-13 root (run 1)	17	Scope 1-6 findings, verification 1-3, triage, partitions, deterministic findings, combined findings, plugin improvements, review report
2026-03-13 run 2	14	Scope 1-6 findings, verification 1-4, triage, partitions, deterministic findings, review report
2026-03-13 run 3	15	Scope 1-6 findings, verification 1-4, triage, partitions, pattern checker, deterministic findings, review report
Combined findings	1	COMBINED-FINDINGS-RUNS-3-4.md
Session 47 reviews	5	Content CRUD, settings/taxonomy, spec, implementation, cross-spec consistency

2.14 .planning/.archive/ — Archived Documents (442 files)

All documents here have been moved from their original locations during the S116/S117 reorganisation. They are kept for decision context and historical reference. All are historic.

Audience: agent | Trust: historic | Currency: historic | Action: keep (archived)

Subdirectory	Count	Contents
.specs/	121	All completed implementation specs — from docs/specs/, .planning/specs/, including verification subdirectory, historical plan documents, session-94 specs, design brainstorm, post-mvp-backlog
.research/	79	Claude capabilities (29), plugin examples (5), tool evaluations (14), market research (5), competitor analysis (3), other research (23)
.continuation-prompts/	73	Session handoffs from S1-S103, s50 redesign plans, s84 fix prompts, standalone prompts
.audits/	70	Session-level audits (S36-S98), s105 audit report and all sub-reports, wp4 verification, gap analyses, critique-gap-crossref, investigation reports, content/data audits, schema/pipeline reviews
.session-exports/	32	S77 transcript + 6 subagent outputs, s91 20 subagent outputs + prompts, session-66b stub, s54 conversation
.claude-desktop/	21	Claude Desktop subagent outputs and feedback files
.session-extracts/	20	Session 36 subagent outputs (17 numbered + main-session + prompts)
.feasibility/	18	Original feasibility study from S34-34b (00-state-of-play through track4-tool-investigation)
.design-audit/	15	Design audit findings — browse, bid-detail, etc.
.extracted-patterns/	8	Extracted patterns — drawer-panel-system, loading-states-pattern, etc.

2.15 .claude/checks/ — Quality Checks (10 files)

File	Purpose	Audience	Trust	Currency	Action
accessibility.md	WCAG 2.1 AA compliance checks	agent	authoritative	current	keep
architecture.md	Architecture pattern checks	agent	authoritative	current	keep
design-system.md	Warm Meridian design system adherence	agent	authoritative	current	keep
error-handling.md	Error handling pattern checks	agent	authoritative	current	keep
image-quality.md	Image quality and optimisation checks	agent	authoritative	current	keep
multi-user-patterns.md	Multi-user and RLS pattern checks	agent	authoritative	current	keep
package-manager.md	Package manager (bun) checks	agent	authoritative	current	keep
supabase-patterns.md	Supabase usage pattern checks	agent	authoritative	current	keep
testing.md	Testing pattern checks	agent	authoritative	current	keep
uk-english.md	UK English spelling and date format checks	agent	authoritative	current	keep

2.16 Project Root (1 file)

File	Purpose	Audience	Trust	Currency	Action	Feeds Into
CLAUDE.md	Project instructions for Claude Code — commands, architecture, gotchas, conventions	agent	authoritative	current (S116)	keep	All areas

---

3. Recommendations

3.1 Documents That Need Updating (Priority Order)

Document	Issue	Effort
docs/reference/ai-integration-strategy.md	Tool counts, sprint status may be stale (last substantive update S82)	Medium — major doc
docs/reference/ai-integration-layers.md	Tool counts, “what’s built vs planned” section may be stale	Medium
.planning/codebase/STRUCTURE.md	File counts and directory listing do not reflect S109-S116 additions and archive reorganisation	Low — can be regenerated
.planning/codebase/TESTING.md	Test count (4,183 in doc vs 6,418 actual) and file counts stale	Low
.planning/codebase/CONCERNS.md	Some concerns may have been resolved in S109-S116	Low
docs/design/warm-meridian-implementation-spec.md	Token map is foundational but may not include tokens added S100+	Low
docs/operations/production-setup-guide.md	May not reflect current Vercel/Supabase config	Low
docs/operations/e2e-test-flows.md	E2E now has 11 spec files; may not cover newer flows	Low
docs/reference/mcp-app-build-guide.md	MCP App architecture may have evolved since S72	Low
docs/reference/state-of-the-product.md	Last updated S107; S108-S116 added significant features	Medium

3.2 Observations

1. The archive reorganisation is a major improvement. Active documentation has been reduced from ~557 files to ~167 active files. This makes it far easier for agents to find what matters.

2. The archive uses dot-prefixed subdirectories (.specs/, .audits/, etc.) inside .planning/.archive/. This keeps them visually separated in directory listings.

3. Continuation prompts now only show recent sessions (S104-S117). The 73 older prompts are archived but still accessible if needed for cross-session archaeology.

4. docs/specs/ is now focused. The 9 remaining files are genuinely active or recently relevant specs, not a mix of completed and active work.

5. Auto-generated docs remain the most trustworthy for volatile numbers. docs/generated/codebase-stats.md and docs/generated/mcp-inventory.md should be the canonical source for tool counts, route counts, component counts, and test counts.

6. Client-facing documentation is well-maintained. The Platform Overview was refreshed to S191 state (23/04/2026). The Claude Integration Guide was last updated in S107 and is pending a separate refresh.

---

4. Product Functionality Mapping

Which existing documents feed into each of the 7 docs/product-functionality/ areas.

4.1 Search

Document	Relevance
docs/operations/search-evaluation-guide.md	Search evaluation methodology and 24 test cases
docs/audits/design-critique-session-4.md	Tom persona critique of search page
docs/reference/classification-prompt.md	Classification drives search relevance

4.2 Content Management

Document	Relevance
docs/specs/upload-page-review-spec.md	Upload page improvements (S107)
docs/operations/bid-library-import-guide.md	Import pipeline operations
docs/audits/design-critique-session-2.md	James persona critique of browse/item detail
docs/audits/design-critique-session-5.md	All personas critique of create content/upload
.planning/.archive/.reference/content-readiness-review.md	Content quality analysis (historic but informative)

4.3 Bid Management

Document	Relevance
docs/specs/bid-session-coverage-spec.md	Bid session enhancements (active spec, B2/B3 remaining)
docs/specs/template-driven-completeness-spec.md	Template completeness (active spec)
docs/specs/bid-testing-strategy-spec.md	Bid UAT scenarios (active spec)
docs/uat/uat-scenario-1-results.md	UAT Scenario 1 results
docs/uat/uat-scenario-2a-results.md	UAT Scenario 2a results
docs/uat/uat-scenario-2b-results.md	UAT Scenario 2b results
docs/uat/s77-uat-research-synthesis.md	UAT research synthesis
docs/audits/design-critique-session-6.md	All personas critique of bid detail/session
docs/reference/product-differentiation-audit.md	Competitive positioning for bid features

4.4 Quality and Governance

Document	Relevance
docs/audits/design-critique-session-3.md	Rachel persona critique of review queue/coverage
docs/reference/product-backlog.md	Main product backlog (renamed from critique-product-backlog.md in S150)

4.5 Knowledge Organisation

Document	Relevance
docs/product-functionality/knowledge-organisation/technical.md	API routes, components, hooks, DB tables, MCP tools (S188)
docs/product-functionality/knowledge-organisation/user-journeys.md	User journeys for all knowledge-organisation subsystems (S188)
docs/product-functionality/knowledge-organisation/workflows.md	Data flows: taxonomy, entity, layer, tag, coverage workflows (S188)
docs/reference/classification-prompt.md	Classification taxonomy (7 domains, 34 subtopics)
.planning/.archive/.research/context-graph-research.md	Context graph research decision

4.6 AI Integration

Document	Relevance
docs/reference/ai-integration-strategy.md	Master AI strategy document
docs/reference/ai-integration-layers.md	Five-layer architecture reference
docs/reference/ai-visibility-policy.md	AI branding and visibility policy
docs/generated/mcp-inventory.md	Current MCP tool/resource/prompt inventory
docs/specs/mcp-evaluation-spec.md	MCP evaluation L5 (active spec)
docs/reference/mcp-app-build-guide.md	MCP App build guide
docs/client-documentation/Knowledge Hub — Claude Integration Guide.md	Client-facing integration docs

4.7 Administration

Document	Relevance
docs/specs/jsonb-normalisation-spec.md	JSONB normalisation (active spec)
docs/operations/production-setup-guide.md	Deployment guide
docs/operations/e2e-test-flows.md	E2E test specifications
docs/operations/e2e-test-setup.md	E2E test data setup
.planning/codebase/CONCERNS.md	Technical debt tracker
.planning/codebase/TESTING.md	Testing patterns and infrastructure
docs/reference/SCHEMA-QUICK-REFERENCE.md	Retired — redirect to generated types

---

5. Key Cross-Cutting Documents

These documents are relevant across all functional areas and should be consulted regularly by authoring agents:

Document	Why It Matters
docs/reference/state-of-the-product.md	Single source of truth for what is built
docs/reference/SCHEMA-QUICK-REFERENCE.md	Retired — redirect to generated DB types
docs/generated/codebase-stats.md	Volatile counts — always current
docs/generated/mcp-inventory.md	MCP tool/resource/prompt list — always current
docs/reference/ai-visibility-policy.md	AI branding policy — invisible infrastructure
docs/design/warm-meridian-philosophy.md	Design principles
docs/design/warm-meridian-implementation-spec.md	Semantic tokens and component patterns
.planning/codebase/ARCHITECTURE.md	Architecture patterns and layers
.planning/codebase/CONVENTIONS.md	Coding conventions
CLAUDE.md	Project instructions, commands, gotchas