Skip to content
cccskills
FreePublished on GitHub

pb-bug-fix

Turn a GitHub issue into a tested fix on a branch with an open PR: read the issue as the contract, find the root cause, write the failing test first, fix, prove it, review the diff against the issue, push and open the PR only after GO. Merging stays with pb-ship. Use for "fix issue 42", "bug fix from this issue", "turn this issue into a PR".

The English summary is being prepared. Showing the original description.

Works withGitHub
See how to install

Files included (1)

  • SKILL.md4.3 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

GitHub issue to tested fix and PR

What you get: a branch with a failing-then-passing test and the fix, reviewed against the issue, and an open PR that closes it. Merging is a separate run (pb-ship).

Inputs

  • repo — the current directory, or the one the user names; <owner/repo> comes from git -C <repo> remote get-url origin
  • issue_number — the GitHub issue to fix, asked if missing

Steps

  1. github — repo, issue_number → gh auth status, then gh issue view <n> -R <owner/repo> --json title,body,labels,state as the contract (acceptance points as a list) in the run log (gh missing, unauthenticated, no GitHub remote, or the issue is closed → log it, give the gh auth login / remote hint, stop) — contract written
  2. deja-memory — error text or symptom from the issue → deja fix result in the run log — prior fix found, or "none"
  3. systematic-debugging — contract → root cause and a local repro command in the run log — repro fails locally on the base SHA (base SHA logged), or the reason it cannot run locally
  4. test-driven-development — root cause → failing test first (run, red, output pasted), then the minimal fix on a new branch fix/<n>-<slug> — the new test fails on the base SHA and passes after the fix; no unrelated edits
  5. verification-before-completion — repro command and the repo's test, lint and typecheck scripts that exist → fresh passing output pasted in the run log — output pasted, not claimed
  6. reviewer (agent) — git diff <base>...HEAD; the contract is the issue from step 1, never the run log and never the implementer's claim (input material only) → findings table in the run log — any open blocking finding sends the run back to step 4, at most 2 cycles, then stop and escalate. No subagent support in this harness → the main session reads agents/reviewer.md and runs the review inline, same contract, findings table in the run log, using only the diff and the contract, never the session's reasoning
  7. git-pr-review — git log <base>..HEAD → PR body draft in the run log, ending in Fixes #<n> — draft only, nothing posted
  8. [GO] git push + gh pr create — git push -u origin <branch> then gh pr create -R <owner/repo> --base <base> --head <branch> --title "<title>" --body-file <draft> — the WAITING FOR GO line names branch, base, title and issue number. The run stops here until the human types GO. One GO covers the push and the PR creation, one time. (Where the hook runs, it covers the push only; gh pr create is not hooked, so this GO is its only guard.)
  9. github — gh pr view <branch> -R <owner/repo> --json url,closingIssuesReferences → PR URL in the run log — closingIssuesReferences contains <n>

Run log: production_artifacts/pb-bug-fix-<date>.md in the start directory, never committed

Rules

  • Anything other than the literal GO (case-insensitive) is not a GO; a GO covers only that one step, one time.
  • GO is by contract in every harness. The go-gate hook is only a backstop on Claude Code, OpenCode and agy (Codex: unverified; Cursor, Kimi: none). A missing hook is never permission to proceed.
  • A failed check stops the run: write the failure into the run log and report. No silent retries beyond what a step names.
  • Write one run-log line per step as it completes (N. done|skipped|failed — artifact — check result) and WAITING FOR GO: <step> at each gate.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Use when manage multiple local CLI agents via tmux sessions (start/stop/monitor/assign) with cron-friendly scheduling.

The English summary is being prepared. Showing the original description.

hybridlabor-api/aos6Updated Oct 16, 2026

Use when a hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).

The English summary is being prepared. Showing the original description.

hybridlabor-api/aos6Updated Oct 16, 2026

Reference for how BDB structures autonomous software engineering work — the seven-node dispatcher graph (Architect, TechLead, UI/UX, Engineering, Media/EventTech, Reviewer, Shipping) that /startcycle-graph actually runs. Use when you need the high-level lifecycle framing without inventing your own process.

The English summary is being prepared. Showing the original description.

hybridlabor-api/aos6Updated Oct 16, 2026

Tools are how AI agents interact with the world. A well-designed tool is the difference between an agent that works and one that hallucinates, fails silently, or costs 10x more tokens than necessary. This skill covers tool design from schema to error handling.

The English summary is being prepared. Showing the original description.

hybridlabor-api/aos6Updated Oct 16, 2026

Harness patterns for coding agents — memory, permissions, context engineering, delegation, skills, hooks, bootstrap.

The English summary is being prepared. Showing the original description.

hybridlabor-api/aos6Updated Oct 16, 2026

agenttrail: live map of a multi-agent build in the browser: which plan component is being worked on, by which agent or harness, what is done and what is stuck. Use when a multi-agent pipeline starts (/startcycle, /startcycle-graph, /teamwork-preview) or after a plan-canvas approve, when the user asks to see what the agents are doing ("live map", "build board", "who is running now"), or by yourself whenever a multi-agent run is under way — no user prompt needed.

The English summary is being prepared. Showing the original description.

hybridlabor-api/aos6Updated Oct 16, 2026

See all skills by hybridlabor-api

Report a problem with this skill