feat: add recommend-new-approaches skill (stage 10)#9
Merged
Conversation
- Add recommend-new-approaches skill (stage 10) — generates prioritized research directions, modeling ideas, and loss function modifications from training artifacts; optionally leverages autoresearch - Wire recommend-new-approaches as step 5d in orchestrate-e2e (always runs after eval/interpretability gates, before promotion decision) - Add recommend-new-approaches JSON schema to references/schemas.md - Add stage-order column to README Available Skills table; split babysit-training (6b) and check-failed-run (6c) as distinct rows - Add CHANGELOG.md mirroring release.yml category conventions - Document changelog update convention in AGENTS.md - Bump plugin version 0.2.0 → 0.3.0
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
recommend-new-approachesskill (stage 10) — synthesizes training artifacts into a prioritized list of research directions, modeling ideas, and loss function modifications; optionally leverages autoresearchorchestrate-e2e(always runs after eval/interpretability gates, informational only)babysit-training(6b) andcheck-failed-run(6c) can be invoked standaloneCHANGELOG.mdwith release.yml-matching category conventions; document update process inAGENTS.md0.2.0→0.3.0Test plan
/recommend-new-approaches --out-dir <dir>after a training run and verify it reads available JSON artifacts and outputs a sorted recommendations listorchestrate-e2eend-to-end and confirm step 5d runs afterexplain-modelwithout blocking the promotion decisionrecommend-new-approaches.jsonis written with correct schema fields (recommendations[]sorted byexpected_impact)--autoresearchflag path logs a warning gracefully whenautoresearchis not installed