cb6d6d46ab
The FactRetriever's _fts_candidates passed the raw query string directly to FTS5's MATCH operator. FTS5 defaults to AND-between-tokens, which means any multi-word prose query like 'what happened with the deployment rollback' required every single token to co-occur in a fact — dropping recall to zero on the kind of queries agents actually issue via prefetch(). Fix: add _sanitize_fts_query() that: - tokenizes the query and drops English stopwords - strips FTS5 operator characters per token - OR-joins the remaining content tokens as phrase literals For pathological inputs (all stopwords, empty), falls back to the raw query so the caller sees zero results instead of a SQL error. This is a pure-retrieval-quality fix — the HRR + Jaccard reranking stages still keep precision high. Ships with 10 tests covering the sanitizer and retrieval integration. |
||
|---|---|---|
| .. | ||
| browser | ||
| dashboard_auth | ||
| image_gen | ||
| memory | ||
| model_providers | ||
| platforms/photon | ||
| transcription | ||
| tts | ||
| video_gen | ||
| web | ||
| __init__.py | ||
| test_achievements_plugin.py | ||
| test_chronos_cron.py | ||
| test_chronos_verify.py | ||
| test_discord_runtime_failure.py | ||
| test_disk_cleanup_plugin.py | ||
| test_google_meet_audio.py | ||
| test_google_meet_node.py | ||
| test_google_meet_plugin.py | ||
| test_google_meet_realtime.py | ||
| test_hindsight_health_grace_timeout.py | ||
| test_hindsight_root_guard.py | ||
| test_kanban_attachments.py | ||
| test_kanban_dashboard_plugin.py | ||
| test_kanban_worker_runs.py | ||
| test_langfuse_plugin.py | ||
| test_nemo_relay_plugin.py | ||
| test_plugin_dashboard_auth_contract.py | ||
| test_raft_check_fn_silent.py | ||
| test_retaindb_plugin.py | ||
| test_security_guidance_plugin.py | ||
| test_teams_pipeline_plugin.py | ||