Bug: "что я говорил про X" escapes the personal boundary and reaches web search #102

2026-08-03T20:57:23+02:00

claude commented

2026-08-03 20:57:23 +02:00

Closes Vikunja #495.

Acceptance criteria and quality gate are in TASK.md on this branch.
Review the review, not the diff — leave comments and the agent will apply them via task start 495.

Closes Vikunja #495. Acceptance criteria and quality gate are in `TASK.md` on this branch. Review the review, not the diff — leave comments and the agent will apply them via `task start 495`.

claude added 1 commit 2026-08-03 20:57:23 +02:00

memory: score the personal boundary on seeds, not word lists (V-495) 86817d6d06

"что я говорил про бэкапы?" is his data by definition, and nothing outside the
box has ever heard him say anything. The boundary matched possession words only,
so the question walked past it into SearXNG and came back answered out of a Habr
article about somebody else's backups.

A speech-verb marker class was written first and dropped. Russian gives every
verb a dozen surface forms and the "как я говорил, ..." preamble list has no end,
so each form the lexicon missed was one more question reaching the world, and a
missing verb looks exactly like no bug.

The boundary now embeds two frozen seed sets and scores the turn's own query
vector, already computed upstream, against both. Nearest side wins. The
possession markers stay as the offline floor for a handler with no embedder.

19/19 held-out utterances correct against multilingual-e5-small; see
docs/evals/2026-08-03-personal-boundary.md. The live probe on the deployed box is
not done.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

kami merged commit 6e5364a0ed into master

2026-08-03 20:59:34 +02:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: kami/Maven#102