Verify extraction provenance and prune semantic staging before inference
This commit is contained in:
1 parent
5c18e7b49c
commit
423a1ea57d
17 files changed
+900
-78
No files matched your search
@@ -30,6 +30,12 @@ readiness without completed release gates and recorded evidence.
|
||||
- Desktop setup must work out of the box: ship the native worker, automatically
|
||||
download/verify missing Google assets, reuse cached assets, and expose progress/retry in the UI.
|
||||
- Add meaningful tests for retrieval, persistence, and filesystem lifecycle changes.
|
||||
- Persist SHA-256 of exact parser input bytes. Same-size/mtime overwrites must
|
||||
be detected on rescans; semantic encoding and agent excerpts must reject
|
||||
missing or stale extraction provenance. Preserve legacy lexical search.
|
||||
- Validate sources after ANN preparation, before the publication write
|
||||
transaction. Prune obsolete staging before inference without dropping
|
||||
compatible completed work or modifying the active generation.
|
||||
- Keep docs/ROADMAP.md current. Unexecuted platform checks remain unchecked.
|
||||
- Gitea Actions are disabled for this repository at Karl's explicit request
|
||||
(2026-10-12, repeated failure emails). Do not re-enable without his request.
|
||||
|
||||
Reference in new issue
Block a user