planning-with-files
Optional persistent file-based planning for multi-phase or long-running work that benefits from recovery across sessions or…
GitHub maintainer queues; issue discovery, triage, and creation; PR creation, walkthroughs, stacking, and babysitting; pending reviews and feedback; CI checks and fixes. Use for read or write work on GitHub issues, PRs, stacks, reviews, or checks.
$ npx -y skills add khaneliman/khanelinix --skill github-toolkit --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/github-toolkitContext preview
The summary Claude sees to decide when to auto-load this skill.
GitHub maintainer queues; issue discovery, triage, and creation; PR creation, walkthroughs, stacking, and babysitting; pending reviews and feedback; CI checks and fixes. Use for read or write work on GitHub issues, PRs, stacks, reviews, or checks.
name: github-toolkit description: GitHub maintainer queues; issue discovery, triage, and creation; PR creation, walkthroughs, stacking, and babysitting; pending reviews and feedback; CI checks and fixes. Use for read or write work on GitHub issues, PRs, stacks, reviews, or checks.
Route to one mode and load only named reference:
1. **issue-creation**: draft or explicitly create issue. Read [issue-creation.md](references/issue-creation.md). 2. **pull-request-creation**: draft or explicitly create pull request. Read [pull-request-creation.md](references/pull-request-creation.md). 3. **pr-stacking**: create, inspect, restructure, or merge dependent pull requests as a GitHub stack. Read [pr-stacking.md](references/pr-stacking.md). 4. **issue-discovery**: search, filter, rank, or summarize many issues. Read [issue-discovery.md](references/issue-discovery.md). 5. **issue-triage**: classify target issue and draft next-step guidance. Read [issue-triage.md](references/issue-triage.md). 6. **pr-review**: review target, keeping audits of the user's own PRs private. For another contributor's PR, prepare a pending GitHub review unless private output is requested. Read [pr-review.md](references/pr-review.md). 7. **pr-feedback**: inspect or address existing review comments. Read [pr-feedback.md](references/pr-feedback.md). 8. **ci-fix**: diagnose failing CI and implement and verify requested fixes. Read [ci-fix.md](references/ci-fix.md). 9. **maintainer-queue**: collect a bounded repository queue, rank evidenced next actions, and route selected items. Read [maintainer-queue.md](references/maintainer-queue.md). 10. **pr-merge**: merge an authorized PR or finish post-merge closeout. Read [pr-merge.md](references/pr-merge.md). Use its closeout after stack merges too. 11. **pr-walkthrough**: explain a PR or prepare a reviewer guide. Read [pr-walkthrough.md](references/pr-walkthrough.md). 12. **pr-babysit**: watch an open PR across wakes and respond to new reviews, bot comments, and checks with local fixes and pending-review drafts. Read [pr-babysit.md](references/pr-babysit.md).
If intent is unclear, ask for mode before GitHub writes or source edits.
template before drafting or publishing.
Audits of the user's own PRs, including PRs created during the task, stay in chat or private reports unless the user explicitly requests a GitHub review draft. Do not create a self-review as a receipt for internal quality checks. For another contributor's PR, a PR review request authorizes creating or updating the current actor's pending review, not submitting it. Explicit chat-only, read-only, or no-post instructions override that default. Other GitHub writes require their own authority; preparing a pending review does not authorize publishing comments, approving, requesting changes, deleting reviews, or resolving threads.
for GitHub Enterprise until hostname binding is implemented.
generic significance claims, canned acknowledgement, or repeated detail.
units. Use pr-stacking mode here to execute those slices as a GitHub stack.
branch deletion. Route verified task-owned local cleanup to `git-toolkit`; honor existing cleanup authority without asking again.
Run packages directly with: Or install from the packages output. For example:
Optional persistent file-based planning for multi-phase or long-running work that benefits from recovery across sessions or…
Audit coding-agent sessions and mine cross-harness history for recurring corrections, skill-routing gaps, and harness faults. Use…
Design and audit agent instructions, skills, plugins, subagents, prompts, hooks, and provider configuration. Use for AI-tool…
Sketch interfaces, state, and module structure before code. Use for explicit design-led implementation or as a caller-owned…
Spawn N parallel candidates at the same task, pick a base, graft the strongest parts of the losers into it. Use for /arena,…
Bevy workflows for ECS, migrations, first-party UI/input/rendering/assets/shaders, plugin selection, scenes, scheduling, hot…