freshness: heal drifted chunks completely — re-embed in place, not deferred (#102)

freshness: stale-while-revalidate — self-heal cited pages, correct the live answer on drift (#101)

retrieval: remove the cross-encoder reranker (ablation showed no earned keep) (#100)

ci: Backfill content workflow (skip crawl) + empty-snapshot purge guard (#98)

perf: store chunk content in Neon so hydration needs no live fetch (Stage A) (#97)

perf: hydrate retrieved sections concurrently — cut the dominant answer latency (#96)

QuOTE questions in the enrichment pipeline + one-click eval default (#88)

Reranker as the precision stage + fold glossing into Build Index (#85)

Extractive synopsis: lead every api chunk with the page's own summary sentence

Widen the HNSW candidate scan: kind-filtered dense queries were dropping true neighbors

Contextual Retrieval: LLM glosses for api reference pages

Fix per-pool relevance gap: a close tutorial no longer empties the api pool

Swap the embedder: bge-small (384d) -> bge-base (768d)

Retrieval redesign: per-kind pools, relevance threshold, interleave

Let the planner choose the content space: search_docs gains a kind filter

Retrieval: one distance-qualified reference seat instead of two blind ones

Retrieval: guarantee reference pages a seat in the top-k

Apply runtime column migrations from the app pool, not only from builds

Cut oversized sections at natural seams; embed only the new parts

Harden the input guard for production

Merge input guard branch (injection + topicality checks)

Serve questions concurrently instead of one-at-a-time (#49)

Add input guard: prompt-injection + topicality check on the user's question

Codebase quality pass: kill silent-truncation hacks, fix latent bugs, cover them (#46)

Fix LLM transport drift, wire live static checks, clean config (#39)

M5: Gradio web app + HF Spaces deploy + live-hydrate fallback

Pin the exact API-reference page to #1 for precise symbol queries

Enrich indexed text with symbol + heading path (give URL/title semantic weight)

Add exact-symbol retrieval channel so API-name queries surface the reference page

M2 retrieval: hybrid search (pgvector + tsvector, RRF) and snapshot hydrate

Switch embeddings to a local CPU model (bge-small): free tier APIs can't index the corpus

M2 indexing: Neon schema, resumable Gemini embedding, one-command build script

Scaffold repo structure per M0 (pyproject, ruff, pytest, pre-commit)