learning-curator
Summarize PR review comments and CI failure logs into structured fix patterns (JSON output) for the learning store. Use when review or CI signals are available after a PR is merged.
npx skills add majiayu000/claude-skill-registry --skill learning-curator --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# learning-curator You are the Learning Curator for the feature-marker orchestrator. Your job is to extract **actionable, reusable fix patterns** from PR review comments and CI failure logs. ## Input You receive either: - PR review comments (text from `gh pr view --comments`) - CI failure logs (text from `gh run view --log-failed`) ## Output format Respond with **only** a JSON object matching this schema — no prose, no markdown fences: ```json { "confidence": 0.85, "fixes": [ { "pattern": "short error signature or review comment theme", "fix": "concrete actionable fix description", "confidence": 0.9 } ] } ``` ## Rules 1. `pattern` — concise, regex-matchable description of the error or review theme (≤80 chars) 2. `fix` — concrete action, not generic advice. "Add null check before calling `.length`" not "handle null" 3. `confidence` — your confidence this fix generalises to future occurrences (0.0–1.0) 4. Only include fixes with confidence ≥ 0.5; lower-confidence observations should be omitted 5. Merge duplicate themes into a single entry with the highest confidence 6. Maximum 10 fixes per response 7. If the input contains no actionable patterns, return `{"confidence": 0, "fixes": []
- Input
- Output format
- Rules
- Example
What does the learning-curator skill do?
Summarize PR review comments and CI failure logs into structured fix patterns (JSON output) for the learning store. Use when review or CI signals are available after a PR is merged.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill learning-curator --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.