Commit Graph

3 Commits

Author SHA1 Message Date
kami 8a174c1c70 Score the recall fixture and write up what it shows
Real recall is 48% after the gate, and one must-be-silent query gets an
answer anyway. Review finding 2 (the score distributions overlap, so no
gate separates a real recall from a false one) and finding 4 (the memStore
branch at voice.go:776 is unreachable for notes). Adds an embedder cache
so the gate sweep does not re-embed the fixture nine times.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CGeSZxh1DCtRxmFVSYVGvJ
2026-07-31 02:34:28 +04:00
kami 43470abc57 Add a held-out note-recall harness (fixture + scorer)
Measures whether Maven can find the right note again from a paraphrased
question. Review internal/memory/recalleval/recalleval.go's Score for how
rank, gate and false recall are kept as three separate numbers, and the
fixture's filler list for why recall@3 is not free.
Fixture JSON is generated data and does not count toward the diff limit.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CGeSZxh1DCtRxmFVSYVGvJ
2026-07-31 02:18:58 +04:00
kami 880715fad4 maven: long-term memory vector-store interface (task 7)
- New internal/memory/ package: Store interface, InMemoryStore (cosine sim)
- Tests: insert→search roundtrip, topK truncation, empty store, cosine edges
- Wire into voice: memStore on reactiveHandler, insert note embedding after WriteNote
- Memory Insert is best-effort, log-and-continue on error

Co-Authored-By: opencode <opencode@anthropic.com>
2026-07-06 04:18:50 +04:00