agentleFS
chuckplayerchuckplayer/claude-agent-pack/skills/hotfix/SKILL.md

hotfix

Skill8 starsChanged 35 days agoMITView on GitHub

Fast-track fix pipeline for production incidents. Skips worktree isolation and planning ceremony. Still requires code-reviewer and merge-reviewer as a safety gate. Use only when the root cause is already known and the fix is well-understood. Trigger this when someone says: emergency fix, production is broken, quick patch, urgent fix, hotfix, critical bug in prod, I know the fix I just need to apply it. Do NOT use when diagnosis is still needed — use /debug first. Do NOT use when the fix touches more than 3 files, adds dependencies, or changes the schema — use /implement instead.

31 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use playtest-report, soak-test and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • CoralGameSkill

    playtest-report

    Generates a structured playtest report template or analyzes existing playtest notes into a structured format. Use this to standardize playtest feedback collection and analysis.

    Also in 12 projects that use this

  • lilangMaxSkill

    soak-test

    Generate a soak test protocol for extended play sessions. Defines what to observe, measure, and log during long play sessions to surface slow leaks, fatigue effects, and edge cases that only appear after sustained play. Primarily used in Polish and Release phases.

    Also in 9 projects that use this

  • lilangMaxSkill

    team-live-ops

    Orchestrate the live-ops team for post-launch content planning: coordinates live-ops-designer, economy-designer, analytics-engineer, community-manager, writer, and narrative-director to design and plan a season, event, or live content update.

    Also in 9 projects that use this

  • CoralGameSkill

    team-polish

    Orchestrate the polish team: coordinates performance-analyst, technical-artist, sound-designer, and qa-tester to optimize, polish, and harden a feature or area for release quality.

    Also in 14 projects that use this

  • +5 more from lilangMax
  • +3 more from CoralGame

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. 0 are common, 12 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 29 edited versions we read keep these.

  • 12

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Before starting, confirm all of the following are true:

    line 14
  2. Very specific:

    If any of these are false, use /implement instead — state which condition failed and why.

    line 21
  3. Very specific:

    If on main or master: create a hotfix/<brief-slug> branch immediately.

    line 27
  4. Distinctive:

    If already on a non-main branch: confirm it is the correct branch for this fix.

    line 28
  5. Distinctive:

    No worktree isolation. The engineer works directly on the hotfix branch.

    line 30
  6. Distinctive:

    If frontend-engineer or mcp-engineer made changes, run ts-linter immediately.

    line 52
  7. Distinctive:

    Always. Pass the changed files and the root cause context.

    line 56
  8. Distinctive:

    If PASS: proceed to step 7.

    line 75
  9. Distinctive:

    If FAIL: fix the blocking issue and re-run steps 5–6.

    line 76
  10. Distinctive:

    After merge-reviewer returns PASS, push the hotfix branch immediately.

    line 80
  11. Distinctive:

    Do not open the PR without these items.

    line 89
  12. Distinctive:

    Scope creep during the fix: If the engineer discovers that the fix requires touching a 4th file, adding a dependency, or changing a schema, …

    line 93
  13. Distinctive:

    merge-reviewer FAIL with no more retries: If merge-reviewer still returns FAIL after 1 retry, do not loop further.

    line 95
  14. Distinctive:

    Diagnosis not yet complete: If you are not certain of the root cause, use /debug first.

    line 97

“Common” is measured: kept by at least 6 in 10 of the 29 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Hotfix

Apply a targeted fix with minimal ceremony. This pipeline is intentionally abbreviated — use it only when the problem is already diagnosed and the change is small and well-understood.

When to use /debug instead: If you don't yet know the root cause, use /debug to diagnose first, then return here to apply the fix.

1. Confirm the fix is in scope for hotfix

Before starting, confirm all of the following are true:

  • The root cause is known
  • The fix touches 3 or fewer files and fewer than 50 lines of logic
  • No new dependencies are introduced
  • No schema changes are required
  • The fix does not introduce a new architectural pattern

If any of these are false, use /implement instead — state which condition failed and why.

2. Branch setup

Run git branch --show-current.

  • If on main or master: create a hotfix/<brief-slug> branch immediately. Do not proceed on main.
  • If already on a non-main branch: confirm it is the correct branch for this fix.

No worktree isolation. The engineer works directly on the hotfix branch.

3. Engineer — direct fix

Route to the appropriate engineer based on file type:

File typeEngineer
.cs / .NETcsharp-engineer
.ts / .vuefrontend-engineer
.ts in an MCP servermcp-engineer
SQL / migrationdatabase-engineer

Pass:

  • Exact files to change
  • Root cause and the intended fix
  • How to verify (test command or manual repro)

No isolation: "worktree" — engineer works directly on the hotfix branch.

4. ts-linter (if applicable)

If frontend-engineer or mcp-engineer made changes, run ts-linter immediately. Block on FAIL before proceeding.

5. code-reviewer

Always. Pass the changed files and the root cause context. This is a non-negotiable safety gate even in a hotfix.

  • Skip security-reviewer and performance-reviewer unless the fix touches auth, data access, secrets, or a critical hot path. State your reasoning.
  • smell-reviewer is intentionally skipped in the hotfix pipeline — structural review of a ≤3-file targeted fix is not worth the latency. This is a deliberate exemption from the "smell-reviewer runs on every code change" rule in CLAUDE.md.

6. merge-reviewer

No plan governs a hotfix, deliberately. This skill passes no plan_id, so merge-reviewer's gate 4a reports "not applicable" and neither the acceptance bars nor the ## Deviations checks apply. That is the exemption working as designed — plan consumption is opt-in per invocation, and a hotfix has no plan because it skipped planning. Do not add plan_id handling here without first deciding who would write the plan and when; an emergency path that acquired a planning artifact would stop being an emergency path.

Pass:

  • Summary of the fix (root cause, what changed, why)
  • Which pipeline stages ran and which were skipped (with reasons)
  • code-reviewer findings

If PASS: proceed to step 7. If FAIL: fix the blocking issue and re-run steps 5–6. Allow 1 retry only — if it still fails, escalate to the user rather than looping further.

7. Push

After merge-reviewer returns PASS, push the hotfix branch immediately. Ask the user whether to open a pull request targeting main and whether to request an expedited review.

test-engineer is skipped in the hotfix pipeline — speed is the priority and the change scope is intentionally narrow.

Required PR description items — the PR body must include both of these before the hotfix is considered complete:

  • **Test coverage gap:** <list each changed method/function that has no test coverage>
  • **TODO:** Add tests for <method(s)> in a follow-up task before next release

Do not open the PR without these items. They create a traceable artifact so the coverage gap is not silently forgotten. If the user declines to open a PR, include the same items in the commit message body.

Gotchas

  • Scope creep during the fix: If the engineer discovers that the fix requires touching a 4th file, adding a dependency, or changing a schema, stop and escalate to /implement. Do not quietly expand scope — the scope gate in step 1 exists for a reason.
  • Skipping code-reviewer: Even under time pressure, code-reviewer is non-negotiable. A hotfix that introduces a new bug or security gap is worse than the original incident. It takes minutes and can catch critical mistakes.
  • merge-reviewer FAIL with no more retries: If merge-reviewer still returns FAIL after 1 retry, do not loop further. Surface the unresolved report to the user and let them decide whether to convert this to a full /implement pipeline.
  • Applying a hotfix to main directly: Step 2 requires a hotfix/<slug> branch even for emergencies. Working directly on main risks losing the change if anything goes wrong. The branch is fast to create and provides an obvious audit trail.
  • Diagnosis not yet complete: If you are not certain of the root cause, use /debug first. A hotfix applied to the wrong place will need to be reverted under pressure — more costly than taking 10 extra minutes to diagnose properly.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.