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>
Types ForkCard/Record/Entry/List, ForkCursor and ForkLimit validated at decode, DeskRequest::Forks
gains status/cursor/limit flat, DeskResponse gains Forks and Fork, the daemon dispatches to
Mind::desk_forks/desk_fork (stubs). Vectors vendored at CultLib e57fbaee. The tests pin the rules
the next commit implements; DESK_READS_UNBUILT and its test are gone.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
SourceQuery gains `front` (serde default false, so an old client's request
decodes as before). The one owner of the list is sources::FRONT_PAGE_FILES;
the Qdrant filter matches `path` against it with MatchAny on the existing
keyword index, composed with the `repo` condition. The CLI prints the
citation line range as before.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
filter and params are asserted by value, so must_not and a top-level exact
fail. live_index_smoke searches a strict subset that leaves out the nearest
point. Unverified at this commit: built and tested once at the end of the batch.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The mind decides membership (Mind::members) before the index is asked; the
index scores exactly the members (Qdrant has_id + exact) and the page carries
one cosine score per row. Deletes OVERSAMPLE_MAX, the x4 oversample, the
unfiltered VectorIndex::search contract and check_semantic.
Not yet compiled when committed: the first Yggdrasil run follows.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Soul pass 2 follow-ups F1, F2, F4, F5 on the semantic index.
F1: every call a search makes (identity reads, label read, embed, query) is
bounded by what remains of one Budget, and the queued ticket carries the moment
SEARCH_DEADLINE after it was asked. The ports take the bound per call
(model_identity, describe, search); QUERY_EMBED_TIMEOUT is gone, since the
ticket's deadline is the one number. A hung identity read frees the worker
within the search's time.
F2: one refusal predicate. Behind is lag and no longer refuses; only
Reconciling, Failing and Refused do, for the sink and the projector alike. The
worker absorbs the inbox before taking a search and leaves the search queued
while the writes ahead of it are unfinished, so a client that admits and then
searches reads its own writes. An index being rebuilt is still refused.
F4: the model-changed refusal and the worker's failure answer to queued
searches name the state and the pending count.
F5: a search that fails for its own reasons no longer forgets the
reconciliation; only a changed model or label does.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Qdrant supplies candidate ids and scores and nothing else. Mind::rank drops
every hit the mind does not hold, then lets CultNet's selection decide
membership over the whole mind with its keys narrowed to the candidates, so
in_force and cited mean what they mean in query. Order is score then ordinal;
the page is cut to the selection's limit and top_k; next is never set. A
cursor with semantic, blank text and an out-of-range top_k are refused by name
before anything is embedded. Mind::query loses its semantic parameter and the
"index is not wired" refusal; the published request schema's description of
SemanticQuery follows.
The serve loop never waits on the embedder: a semantic query becomes a Search
held beside its session, the worker embeds (behind the model's query
instruction) and searches between batches, and the loop answers when the
worker has or when ServeOptions::search_timeout passes. A worker that is
Failing or Refused refuses at once, typed Unavailable naming its own error.
VectorIndex gains search; Qdrant's adapter sends no filter.
Model identity is re-read before every flush and before every search: a model
re-pulled under the same dimensions but a new digest forgets the
reconciliation instead of mixing vectors, and the next step rebuilds. An idle
worker forgets the reconciliation every Backoff::recheck (60 s), so a lost
collection is found without a write. Both go through Projector::reconcile,
still the one judge.
Also fixes a race in the worker loop found by the mutation runs: entries were
taken from the inbox under one lock and published as wanted under another, so
whoami could read a stale Current with nothing queued while the worker held
them. Taking, absorbing and publishing are now one critical section.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
