docs: README §6.2/§4.5 — reflect get_policy_facts grounding, TTS chunking, barge-in abort, interim accumulation

fix(chat): thread AbortController signal into postChat() so barge-in cancels the in-flight request

docs: README — desktop voice is push-to-talk (hold-SPACE was removed)

fix(batch): voice interim, TTS truncation, RULE2.5 re-ask, faithfulness regression, markdown, compare-reviews, policy-name

fix(brain): don't re-ask recalled slots after returning-user confirm

fix(brain): wire returning-user-by-name recall into single_brain (ADR-041/KI-196)

fix(voice/brain): remove hold-space, fix silent TTS + policy-Q&A refusal

docs(#27): README §6.4 Known limitations — honest current-state residuals

feat(#77 / #52-residual-5): admin prune for persisted uploaded docs

fix(#61): named-policy bypass not gated on intent (single_brain hardcodes it)

fix(#61): resolve the version-specific chunk-bearing card, not the empty base

fix(#61): named-policy Q&A bypasses the recommendation profile-gate

Revert "Revert "feat(#52): PDF upload → persistent JSON + marketplace card + restart-survival""

Revert "feat(#52): PDF upload → persistent JSON + marketplace card + restart-survival"

feat(#52): PDF upload → persistent JSON + marketplace card + restart-survival

fix(#63): header premium pill is the TYPICAL cohort range, labelled as such

fix(#65): surface claim-experience on inline chat compare cards

fix(#66): enrich the sparse POLICY DETAILS expander

fix(#76): show the customer's chosen cover, not 'As per policy schedule'

fix(#64): preserve EXACT budget ₹ — slider shows what the user said

fix(#53+#54): page.tsx SPACE-hold PTT delegates to warm-stream hook

fix(#75): compare_policies curated fallback — no scorecard surface 404s

fix(#55+#56+#53/#54-engine): TTS full natural readout + voice warm-stream/pre-roll

fix(#61): expand policy_id retrieval filter to the canonical family

fix(#57+#58+#59): canonical-resolve compare-card marketplace lookup

fix(#60+#62): bulk-scorecard curated fallback + stop internal scoring jargon leaking

fix(#40+voice+upload): SSOT scorecard parity + Sarvam 30s-truncation + uploaded-PDF retrieval gate

fix(pricing+scoring): rebuild premium engine to 0 logical violations + classification, faithfulness, links, writing

fix(reco): cap 3, exclude non-extracted policies, unify premium, ask PED

chore(cleanup): frontend — drop stale narrative comments (current-standard)

chore(cleanup): purge stale narrative/tombstones/dead code — codebase reads as the current standard

fix(retrieval): structural prose⇄card 1:1 gate (KI-PROSE-CARD-1to1)

fix(retrieval): quality-seeded candidates + wide recall pool — surface the genuinely-best policy (KI-QUALITY-SEED / KI-RECALL-POOL)

fix(scorecard): rec-path scorecard == marketplace (KI-FULLFACTS) — restores recommendation cards

fix(single_brain): restore broken core flow — model revert + thinkingConfig + rip deterministic scaffolding

feat(audit): selftest in pytest gate (static-scoped, no recursion) + opt-in git hooks + README

feat(audit): Tier 5 deploy-safety (read-only) + selftest_expect for WARN checks

feat(audit): Tier 4 functionality (API smoke + Playwright E2E) + selftest fixtures

feat(audit): Tier 3 build/test gates + selftest fixtures

feat(audit): Tier 2 code-soundness (corrected spec) + core.selftest hardening

fix(de-stale): eval/run.py --no-extract help no longer names deleted backend.profile_extractor/orchestrator (T2.3 finding)

fix(de-stale): remove deleted get_fast_brain_llm (broke rag.extract import)

feat(audit): Tier 2 code-soundness checks + selftest fixtures

chore(repo-hygiene): fix the 3 issues the audit framework's Tier-1 surfaced

feat(audit): Tier 1 repo-integrity checks + selftest fixtures

feat(audit): core contract + runner + CLI

fix(ui): insurer logos — transparent on card, aspect-correct (not distorted)

docs(plan): error/risk audit framework implementation plan

docs(spec): error/risk audit framework design (awaiting review)

recovery: integrate stalled-session work + de-stale/cleanup (pytest 215 green) [build-fix]