GameCult/Huginn
Watch
9
0
Fork
You've already forked Huginn
 0
Code Issues Pull requests 1 Projects Releases Packages Wiki Activity Actions

Source search takes up to eight texts, fused by reciprocal rank (k=60) #52

Merged
eureka merged 5 commits from gamecult-chat/source-query-texts into main 2026-11-11 16:20:41 +00:00
Conversation 0 Commits 5 Files changed 11 +344 -30
eureka commented 2026-11-11 16:09:51 +00:00
Owner
Copy link

Executes cut spec gamecult-chat:cut_spec:cut-source-query-texts.r1 (map cut 1, multi-text-query, docs/chat-retrieval-map.md).

SourceQuery gains texts (one to eight); sources::search embeds all distinct trimmed texts in one embed call, searches once per vector (FUSE_DEPTH = 50 deep when several), and fuses with reciprocal rank fusion (FUSE_K = 60, Cormack, Clarke, Buttcher, SIGIR 2009) through a pure generic fuse the chat cut reuses. One text answers exactly as before (cosine scores, top_k deep, unfused). text stays as a transitional field, documented as such, to be deleted once no client sends it. huginn sources <TEXT>... takes one to eight positional texts and always sends texts; the daemon and CLI deploy together.

Commits, in the spec's order: (1) wire fields, door, regenerated request and response schemas; (2) fuse and its tests; (3) search over the batch, with the behavioural tests; (3b) daemon door test; (4) CLI.

Verification: the Yggdrasil verify path (ygg-verify.sh) is unreachable from the Self body (ssh: Could not resolve hostname ygg), so the focused suites ran locally in a scoped target dir: huginn-mind 415 passed, huginn-daemon lib 205 passed and its other targets green, eureka-state cli 80 passed; cargo check --locked --workspace exit 0. One failure, eureka-state tests/mcp.rs::selection_schema_matches_cultlib, needs a network crate download under the Self's offline cargo and does not touch this diff. No mutation pass was run (needs the verify path).

Not merged, not deployed.

🤖 Generated with Claude Code

Executes cut spec `gamecult-chat:cut_spec:cut-source-query-texts.r1` (map cut 1, multi-text-query, `docs/chat-retrieval-map.md`). `SourceQuery` gains `texts` (one to eight); `sources::search` embeds all distinct trimmed texts in one `embed` call, searches once per vector (`FUSE_DEPTH` = 50 deep when several), and fuses with reciprocal rank fusion (`FUSE_K` = 60, Cormack, Clarke, Buttcher, SIGIR 2009) through a pure generic `fuse` the chat cut reuses. One text answers exactly as before (cosine scores, top_k deep, unfused). `text` stays as a transitional field, documented as such, to be deleted once no client sends it. `huginn sources <TEXT>...` takes one to eight positional texts and always sends `texts`; the daemon and CLI deploy together. Commits, in the spec's order: (1) wire fields, door, regenerated request and response schemas; (2) `fuse` and its tests; (3) `search` over the batch, with the behavioural tests; (3b) daemon door test; (4) CLI. Verification: the Yggdrasil verify path (`ygg-verify.sh`) is unreachable from the Self body (`ssh: Could not resolve hostname ygg`), so the focused suites ran locally in a scoped target dir: `huginn-mind` 415 passed, `huginn-daemon` lib 205 passed and its other targets green, `eureka-state` cli 80 passed; `cargo check --locked --workspace` exit 0. One failure, `eureka-state` `tests/mcp.rs::selection_schema_matches_cultlib`, needs a network crate download under the Self's offline cargo and does not touch this diff. No mutation pass was run (needs the verify path). Not merged, not deployed. 🤖 Generated with [Claude Code](https://claude.com/claude-code) 
SourceQuery gains `texts` (one to eight, SOURCE_TEXTS_MAX) beside the
transitional `text`; check() refuses neither, both, more than eight and blank
entries, naming `texts`. SourceHit.score documents cosine versus reciprocal-rank
sum. Published request and response schemas regenerated. Cut
gamecult-chat:cut_spec:cut-source-query-texts.r1, commit 1.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
fuse is a pure generic function over ranked lists and a key closure, with
FUSE_K = 60 (Cormack, Clarke, Buttcher, SIGIR 2009) and FUSE_DEPTH = 50 named
beside CORPUS. Nothing calls it yet. Cut
gamecult-chat:cut_spec:cut-source-query-texts.r1, commit 2.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
search() embeds each distinct trimmed text (Instruct:-prefixed) in one batch,
searches once per vector, and fuses with reciprocal rank (k=60, depth 50) only
when more than one text remains; one text is searched top_k deep and answers
with the cosine scores as before. Cut
gamecult-chat:cut_spec:cut-source-query-texts.r1, commit 3.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Zero, nine and blank texts answer Refused naming `texts` and never reach the
index; two texts are one queued search. Cut
gamecult-chat:cut_spec:cut-source-query-texts.r1.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The CLI sends every positional text as `texts` and never `text`; help says each
is a topic and the lists are fused. Daemon and CLI deploy together. Cut
gamecult-chat:cut_spec:cut-source-query-texts.r1, commit 4.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
eureka merged commit b373c2a5b1 into main 2026-11-11 16:20:41 +00:00
Sign in to join this conversation.
Reviewers
No reviewers
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
GameCult/Huginn!52
No description provided.