Menu
Apply the diff/patch outputs from a /mine-transcripts run to lessons.md and rules/*.md after human review. Use after running /mine-transcripts and reviewing the proposed diffs. Triggers on /apply-mining-patches, "apply the mining patches", "apply mining run", "approve and apply the trajectory diffs".
| name | apply-mining-patches |
| description | Apply the diff/patch outputs from a /mine-transcripts run to lessons.md and rules/*.md after human review. Use after running /mine-transcripts and reviewing the proposed diffs. Triggers on /apply-mining-patches, "apply the mining patches", "apply mining run", "approve and apply the trajectory diffs". |
| when_to_use | After /mine-transcripts produces diffs and you've reviewed them. Skips approved-then-undone work. |
| allowed-tools | Bash, Read, Write, Edit |
| arguments | ["run-timestamp"] |
/apply-mining-patches <run-ts> applies the lessons + rules diffs from ~/.claude/telemetry/mining-runs/<run-ts>/ to live files, after the user has had a chance to review.
<run-ts> is "latest", glob ~/.claude/telemetry/mining-runs/2* and pick the most recentRead and summarize:
manifest.json — sessions mined, signal countslessons.diff — proposed additions to ~/.claude/lessons.mdagent-tuning.md — agents with ≥2 occurrences worth tuningrules-<name>.diff filesPresent a summary like:
Mining run 20260510T120000Z:
- 12 sessions, 7 corrections, 3 patterns, 4 failure modes
- lessons.diff: +45 lines (3 new lesson entries)
- agent-tuning.md: 1 agent recommended (flake-detector, 4 occurrences)
- rules-self-improvement.diff: +8 lines
Apply? (yes/no/preview)
If preview, show the actual diff content (truncated). If no, exit.
For lessons.diff: append to ~/.claude/lessons.md.
cat <(echo) <(cat <run-dir>/lessons.diff) >> ~/.claude/lessons.md
For rule diffs: append to the named rule file.
For agent tuning candidates: do NOT apply automatically. Surface the /tune-agent <name> command and let the user run it explicitly. Reflexion patches require their own approval.
After successful application, write to <run-dir>/applied.marker with the apply timestamp. This prevents accidental double-application.
If the user wants, commit the lessons/rules updates with a message like:
docs(memory): apply mining run 20260510T120000Z
3 new lessons, 1 rule update from 12 sessions of trajectory mining.
Refs: ~/.claude/telemetry/mining-runs/20260510T120000Z/
Run an agent N times in parallel against the same prompt, then aggregate via one of 5 modes — critic argmax, USC consistency, confidence-weighted, hybrid, or AggAgent synthesis. Use for high-stakes invocations where you'd rather pay Nx to be sure. Triggers on /best-of-n, "best of n", "run multiple", "give me three options", "synthesize across rollouts".
Run a complete SCRUM sprint with all ceremonies — Product Owner authors stories, Tech Lead designs, Scrum Master orchestrates implementers, Verifier+Aspect-Panel guard quality, Sprint Auditor adversarially audits, Retro feeds back to /tune-agent. Triggers on /scrum, "run a sprint", "scrum me this", "ship this with full process".
Run an agent N times in parallel against a code-change task, with each rollout in its own sandboxed copy of the codebase, scored by ACTUAL test pass rate (not just critic opinion). Use for SWE-bench-style code changes where the test suite is the ground truth. Triggers on /exec-grounded, "execution-grounded", "verify by running tests", "best-of-N with tests".
Mine past Claude Code session transcripts (JSONL) to extract user corrections, successful patterns, and recurring failure modes. Proposes diffs against lessons.md and rules/*.md for human review. Triggers on /mine-transcripts, "mine my sessions", "what did I learn this week", "session retro". Use weekly or after a fleet completes.
Verify UI changes by capturing before/after screenshots and asking Claude vision to judge whether the change matches intent and didn't break anything else. Mano-verify pattern (arXiv 2509.17336 —
A/B test an agent's current prompt against a candidate variant from policy/<agent>/candidates/. Runs both on the same scenarios, aggregates rewards, recommends promotion if candidate beats current by >1 stderr with n≥10. Triggers on /ab-test, "compare prompts", "test the candidate", "is the new prompt better".