Repository navigation
Admit bounded indexed diagnostic evidence before compute - #108
Merged
Merged
Conversation
added 3 commits
October 19, 2026 22:11
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars. ⚙️ Run configuration
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
MrScripty
changed the base branch from
feature/lab-phase-panel-20261019
to
feature/lab-batch-client-20261019
October 19, 2026 23:02
added 7 commits
October 19, 2026 23:02
…exed-diagnostics-20261019
MrScripty
changed the base branch from
feature/lab-batch-client-20261019
to
main
October 20, 2026 03:15
MrScripty
marked this pull request as ready for review
October 20, 2026 04:27
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds opt-in bounded indexed scientific diagnostics: the SDK writes an index and shards after scientific output, the host verifies their declared bounds and hashes, and cold readers inspect them through existing artifact and batch interfaces. Existing canonical evidence semantics and the original Run outcome remain intact.
Writer, publication-recovery and composed-reader qualifications retain their recorded source identities and limits: these fixtures use synthetic terminal arrays and non-runnable runtime metadata; they do not qualify physical validity or captured-Python training. The missing original PUB-1 failure bytes and SCI-1 owned criterion/compatible pair remain explicit.
Current head
9f08427029d8b48c8f9b50927a1c835afb573276repairs the sole full-suite failure: an authoring test compared a frozen canonical SDK with the evolving live SDK. The test now verifies the existing exact pinned SDK closure and authored-copy bytes. Frozen artifacts, source-authority hashes and product code are unchanged. The original failure was reproduced; 16 targeted tests passed without skips, with exact owned-schema cleanup. Independent review is clear. Exact-head lifecycle run 39360311688 passed all 172 steps: 1,744 Python tests (72 skipped), scientific/coordinator/phase browsers, actual CPU reconstruction, finite model development and actual text/audio one-fit gates. Independent completed-log review confirms the executed checkout/source attribution. The tested historical base tree equals current main63f689ec; serial integration verifies the prospective and actual merged trees rather than relabelling the tested history.