enne2
|
128059dd4b
|
feat(gateway): strategie rerank oltre la search (A-F)
- A: gate store con cross-encoder — guardrail.decide async, conferma/scarta
quasi-duplicati (CROSS_DUP_CONFIRMED/WEAK/LOW_COSINE), suggerimento
supersedes in WARN, degrada a cosine-only se il reranker è giù
- B: verifica supersede — cross-score (nuovo,vecchio) sotto soglia →
supersede_warning non bloccante + audit
- C: score composito in search — rerank + importance (nuovo campo payload)
+ recency decay (180gg) + authority, pesi SCORE_W_* da env
- E: multi-query — SearchIn.queries (max 3), pool unito con dedup, rerank
unico; endpoint POST /v1/score come primitiva cross-encoder (F-lite)
- extension search.ts: param queries + rerank_score/composite in output
- D: scripts/consolidate.py — dedup periodico a coppie via cross-encoder
con report ntfy e --apply via gateway
- test: 69 pass (+11 strategie); guardrail_version similarity-v2
|
2026-09-08 12:57:33 +02:00 |
|
Matteo Benedetto
|
4776b4b496
|
refactor: split gateway and pi-qmem extension into modules
- gateway: separate config, models, state, audit, guardrail, embeddings,
store, metrics, cleanup and routes; keep main.py as FastAPI bootstrap
- extension: split client/config, six tools, config command and rules;
preserve jiti entrypoint and registrations
- Dockerfile copies the complete gateway module set
- tests: update monkeypatch boundaries for modular config/state
|
2026-08-24 16:19:19 +02:00 |
|