S11 applied with the sequential check off and seen to fail on it. Finding desk-wire-replay-id-map-not-injective.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Soul's S12 (a second converter that agrees on every legal key) applied and seen to fail. Finding desk-wire-second-converter-evades-grep.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Padded, cased and trailing-junk spellings added to the existing tables; Soul's S1, S8, S9, S10 each applied and seen to fail. Finding desk-wire-pass-through-unpinned.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Kills Soul's S2, S3 and S5 (each applied and seen to fail). Finding desk-wire-asof-and-gate-thinly-pinned.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Her vectors hold no over-long fork id (deviation recorded). Finding desk-wire-long-fork-id-body-invalid.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
answer_desk's (answerer, request_id) lookup is now bound_answer, pub(crate) in desk.rs, and
Mind::desk_answer asks it so the gate runs only when nothing is stored. Finding
desk-wire-gate-shadows-replay.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Vendors fork-card-vectors.json at CultLib 8ca1d184 and retires the options-hash file it contains.
answer_desk's dead-code allowance goes: desk_answer is its live caller. No card is built here.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
body_digest writes retracted for the illegal form only, so forms (i) and
(iii) no longer share a digest; desk_steps refuses a value that is no
illegal form as body invalid. Tests pin the three illegal goldens, the
200-byte target, choice and Refused.retracts bounds, case-sensitive
labels and fork_closed before bad_retraction.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
PipelineFacts::DeskAnswer gains counts, derived through counted_answer over the
read's Docs::at(image, as_of); the stored outcome name stays the write-time record.
DeskOutcome::Refused carries the retraction target as sent; the record's choice is a
Short kept as sent (a choice_invalid choice is no label), and an illegal retraction
form keeps its choice and note. body_sha256 form unchanged. Schemas regenerated.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A bound (answerer, request_id) replays its stored outcome without re-running
probe validation; a malformed id binds nothing and still meets the probe.
Comment on the choice bound now says a Short (200 bytes).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
DeskBody::IllegalRetraction stores Refused bad_retraction after steps 4-6 (digest carries the form, so it never replays as a plain answer); a choice is any 1-200 byte text, stored Refused choice_invalid when no option label, grammar or not. claim_in_future compares against the whole-second received_at written to the record.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
DeskStamp carries answerer, attribution (Operator/Steward/Member per CultLib PR #72) and the decider
list; membership decides rule 5, the attribution never does. answer_desk and DeskStamp are crate-private.
The receipt names the operator only for the operator's account. counted_answer and R1 consider
options_hash. A stored answer keeps received_at, claim_in_future, request_id and a body digest;
the same (answerer, request_id) and body replays the stored outcome and writes nothing, another
body is request_id_conflict. Notes are Words (1..=4000 bytes, not blank); answered_at must be UTC.
The vector file is re-sliced byte for byte from PR #72 head 2dea6c56.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
NFC every string, JCS (struct fields in key order through serde_json), SHA-256,
lowercase hex; the seam stays the single owner. Libby's 15 vectors, A-G included,
run as tests; the test fixture hash is gone. Adds unicode-normalization.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Rules 3-8, R1, R2, R4 of the binding spec; counted_answer derives which answer
counts; Door::Desk is the only door that may write a desk_answer, so Admit,
import and derivation refuse it. options_hash sits behind one seam, failing
closed until the bytes are filled; tests use a named fixture hash.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
One kind, key <campaign>:desk_answer:<question label>.n<sequence>, one tagged
outcome (Counted, Revision, Retraction, Refused). Published schema, index row,
golden line and the PipelineKind enum in sibling schemas regenerated.
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>
One gamecult.persona.chat_read.v1 [persona, room, event_id, at_unix_ms] per room per echoed batch,
sent over the same RUDP endpoint as chat_ack, from PersonaInbox::echoed (the CLI has taken the line).
Newest placed event only; no event ids on earlier lines. Display-only: a failed report is dropped.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The skip list now matches `license`, `licence`, `copying` only with no extension or md/txt/rst, so src/license.rs and Copying.cs stay indexed. CHUNKER_VERSION is back to 1: a bump rebuilds the live collection; the already-indexed noise paths go by a one-time delete_by sweep instead (work/sweep/sweep-sources-noise.py, same rule).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The journal line carries the full result; a result longer than a reply's
bound is emitted as consecutive replies (-r{n}), cut at character
boundaries, with the same answers. RESULT_TEXT_LIMIT and the
[truncated] marker are deleted.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A follow-up the receiver wrote natively, closed, and a later batch replays is inert to the
daemon now and refused as an identity collision when the batch arrives.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
inert_in_window covered FollowUp and Idea only, so a receiver's own resolution
of such a record, admitted after the import began, refused the rerun in both
the tool (resume/window_writes) and the daemon (root_holds_native_write).
A Resolution whose subject is a follow-up or idea the plan does not replay
touches nothing planned and now stays beside the replay. One predicate, both
callers; the resolution of a replayed follow-up, or of any other kind, still
refuses.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A closed row is corrected by withdrawing its resolution and resolving the
subject again at the next sequence; admission already allows it, in one batch
too (new test). The refusal gave no way to find it: a Mind that tried to
Withdraw or re-resolve the follow-up itself got AlreadyResolved with only the
subject. It now carries closed_by, the resolution key to withdraw.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
