Repository navigation
feat(lore): review advisory skill evaluation evidence - #263
Merged
Merged
Conversation
Pin LORE 0.62.24 and show pending coding-skill experiment results alongside the exact proposal. Label stale baseline/proposal evidence explicitly while retaining raw-byte digests and the human approval gate. Count the full evidence display in the existing read-to-end guard. Cover old carriers, native unevaluated skills, stale reports and malformed sidecar metadata; keep task evaluation an explicit human-terminal operation.
Replace the reviewed candidate revision with the canonical release commit. The upstream release tree is byte-identical to the verified candidate; link the versioned evaluation workflow.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Pending skill proposals currently display no task-level evaluation evidence. Pin the new LORE experiment workflow and show its advisory results alongside the exact proposal, with an explicit stale label when the proposal or installed baseline changed.
Count the evidence section in the existing read-to-end approval guard. Preserve the raw proposal digest and human approval authority, and remain compatible with older carriers that omit evaluation metadata. Evaluation itself stays an explicit human-terminal operation.
Pins the merged LORE 0.62.24 release, commit
d0dcbcc9a06690a1c340c858c5590d23fe814b77. Its tree is identical to the verified upstream candidate.Validation in an isolated committed-tree checkout: workspace compilation, the full doxa-lore native/sidecar suite, 24 LORE-related TUI tests and the read-to-end approval regression passed. The canonical-release pin recheck also passed; CI runs the broader Linux/macOS/Android gates.
The supplied upstream coding fixture validates the mechanism. It does not measure a real model; no automatic experiments or skill installation are added.