fix(link-rot): validate PDF candidates + untrack llm_health.json

test(#44): add SSOT regression test for canonical_decision_facts

docs: Cluster A (count drift) + Cluster B (deleted-module refs) sweep

docs(README): fix mermaid sequenceDiagram parse error in §3.7

docs: stale-references sweep — align legacy docs with ADR-044 + 8-gate state

docs(README+CLAUDE): wire in KI-332 multi-pass + heuristic-floor expansion

fix(upload-extract): status parity on LLM-fail path + docs updated

feat(upload): multi-pass per-section extraction + heuristic-floor expansion (KI-332)

fix(upload): show heuristic-floor card on LLM-fail path + extend poll to 150s (KI-331)

feat(single_brain): wire view_context → ACTIVE POLICY DIVE-IN block (KI-330)

fix(upload-extract): hash-cache short-circuit now matches scorecard endpoint

fix(upload-extract): read sc.grade not sc.overall_grade

fix(upload-extract): use _catalogue_scorecard primary path for status

fix(upload-extract): status completeness now matches scorecard endpoint

chore: surface llm_used + llm_response_chars on extraction-status endpoint

feat(upload-extract): Tier-2 content-hash cache — same PDF, no re-extraction

feat(upload-extract): Tier-1 Gemini stability + new ADR-044 + README rewrites

fix(upload-extract): drop responseMimeType=application/json — was producing unparseable payloads

fix(upload-extract): LLM output now merges INTO heuristic record (not replace)

fix(upload-extract): Gemini-primary done right — max_tokens 2048→8192 to stop truncation

revert(upload-extract): roll back Gemini-primary switch — surfaced as live failures

feat(upload-ux): lock canonical sequence — ack → wait (everything gated) → card → choice

feat(upload-extract): switch primary LLM from NIM chain → Gemini 2.5-flash (JSON mode)

hotfix: import Header from fastapi — caused RUNTIME_ERROR on 24349be

fix(voice): gate the PTT/Sarvam send-paths on uploadStatus + extractionInFlight too

feat(upload): backfill LLM extraction for old uploads + admin re-extract endpoint

fix(upload): missing asyncio import — module-level Lock() fired NameError on import

feat(upload): wait-for-extraction flow + status endpoint + extended voice guard

feat: uploaded-PDF parity with catalogued 148 (ADR-044)

fix(frontend): PDF upload — in-chat ack + inline scorecard card + proceed-choice prompt

feat!: remove cross-session profile recall (ADR-043) — net −3700 LOC

fix(privacy v3/v4): two-fact recall gate + extended same-turn extractors (closes ADR-042 follow-ups)

docs: ADR-042 + README + CLAUDE.md updates for the 4-fix bundle

fix(privacy v2): same-turn age contradiction in apply_pending_recall

fix: 4 hardening fixes — sticky-retry, canned-msg copy, admin refresh wiring, recall privacy

fix(README): unbreak §3.5 / §3.6 mermaid (HTML entities in participant labels)

docs(README): §2 four-level abstraction + §3 sequence diagrams + dedup §4

docs(README): major rewrite per product+fintech expert review (May 25)

docs(README): plain-language explanation under every Mermaid diagram

docs: README/ARCHITECTURE — list get_policy_facts tool + note #47 UIN dedup

#41 + #43: fix family-history re-ask + unverifiable-UIN warning

#47: UIN-based net-new dedup for user-uploaded policy PDFs

docs: add §4.1 plain-English user-journey diagram; renumber 4.x

docs: expert README polish — rich Mermaid architecture diagrams, clearer data-split table, collapsible repo map

#44: reconcile PED waiting across policy-facts doctype files (data integrity)

#45: persist chat-only profiles so returning-user recall actually works

#25/#26: returning-user recall any-turn + mid-conversation state recovery

#33: add "?" help affordance to every policy-calculation control

feat(#31): deterministic profile-aware {strengths,caveat} policy summary on all 3 surfaces

fix: deductible gating (#29), existing-cover steering + dangling-turn guard (#30), compare-modal full reviews (#32)