How do I install the Apply Mining Patches skill?

Run npx skillmds@latest add sethdford/apply-mining-patches in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Apply Mining Patches safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Apply Mining Patches?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Apply Mining Patches free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Apply Mining Patches?

sethdford (@sethdford) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Apply Mining Patches by sethdford — Agent Skill

Name: Apply Mining Patches
Author: sethdford

Apply Mining Patches

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".

sethdford Updated Sep 17, 2026

File contents

/apply-mining-patches — Apply approved trajectory-mining diffs

/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.

What you do

1. Resolve the run

If <run-ts> is "latest", glob ~/.claude/telemetry/mining-runs/2* and pick the most recent
Otherwise use the exact timestamp directory

2. Surface what's there

Read and summarize:

manifest.json — sessions mined, signal counts
lessons.diff — proposed additions to ~/.claude/lessons.md
agent-tuning.md — agents with ≥2 occurrences worth tuning
Any rules-<name>.diff files

3. Confirm with user

Present 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.

4. Apply diffs

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.

5. Mark applied

After successful application, write to <run-dir>/applied.marker with the apply timestamp. This prevents accidental double-application.

6. Commit (if in a git repo)

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/

Anti-patterns

Applying without showing the user the actual content first — patches can be wrong
Auto-applying agent-tuning candidates — those need /tune-agent's own approval flow
Applying a run that's already marked applied — fail loudly
Discarding a run's directory after apply — keep it for reproducibility
Editing the original miner output file — the diffs are immutable; if a diff is wrong, that's a bug in the miner, fix the miner

---name: apply-mining-patchesdescription: 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".---# /apply-mining-patches — Apply approved trajectory-mining diffs`/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.## What you do### 1. Resolve the run- If `<run-ts>` is "latest", glob `~/.claude/telemetry/mining-runs/2*` and pick the most recent- Otherwise use the exact timestamp directory### 2. Surface what's thereRead and summarize:- `manifest.json` — sessions mined, signal counts- `lessons.diff` — proposed additions to `~/.claude/lessons.md`- `agent-tuning.md` — agents with ≥2 occurrences worth tuning- Any `rules-<name>.diff` files### 3. Confirm with userPresent 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 linesApply? (yes/no/preview)```If preview, show the actual diff content (truncated). If no, exit.### 4. Apply diffsFor lessons.diff: append to `~/.claude/lessons.md`.```bashcat <(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.### 5. Mark appliedAfter successful application, write to `<run-dir>/applied.marker` with the apply timestamp. This prevents accidental double-application.### 6. Commit (if in a git repo)If the user wants, commit the lessons/rules updates with a message like:```docs(memory): apply mining run 20260510T120000Z3 new lessons, 1 rule update from 12 sessions of trajectory mining.Refs: ~/.claude/telemetry/mining-runs/20260510T120000Z/```## Anti-patterns- Applying without showing the user the actual content first — patches can be wrong- Auto-applying agent-tuning candidates — those need /tune-agent's own approval flow- Applying a run that's already marked applied — fail loudly- Discarding a run's directory after apply — keep it for reproducibility- Editing the original miner output file — the diffs are immutable; if a diff is wrong, that's a bug in the miner, fix the miner

Frequently asked questions