feat: improve agent migration skill with measured evals#81
Draft
skillets-the-skill-intern[bot] wants to merge 1 commit into
Draft
feat: improve agent migration skill with measured evals#81skillets-the-skill-intern[bot] wants to merge 1 commit into
skillets-the-skill-intern[bot] wants to merge 1 commit into
Conversation
Co-Authored-By: Claude <[email protected]>
Contributor
Author
|
I mistakenly ran this improvement through Skillet’s legacy |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Improves
openrouter-agent-migrationusing a fixed, hermetic before/after evaluation suite.Measured result: run-level success improved from 6/9 (66.7%) to 9/9 (100%); expectation-level pass rate improved from 108/111 (97.3%) to 111/111 (100%). The difficult streaming-converter fixture improved from 0/3 to 3/3 without unsafe casts or lint/type suppressions.
What changed: evidence-driven dependency compatibility checks, agent-vs-platform inventory, safe converter-mismatch handling, mock and lockfile migration guidance, and verification for stale imports, suppressions, tests, type checking, and minimal diffs.
Eval suite: three pure TypeScript fixtures with mocked tests and deterministic validation; no live API calls.
Tradeoff: final safety-oriented runs were slower, especially converter migrations, because they validate compatibility and avoid suppression-based shortcuts.