菜单
Automated dev loop — run tests, file issues for failures, TDD fix on branch, open PR, watch CI. Use when running a continuous dev loop, autonomous TDD, or fix-and-PR cycles.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/laurigates/claude-plugins --skill workflow-dev命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Install, configure and troubleshoot MCP servers. Use when adding/enabling servers, editing .mcp.json, fixing OAuth, or when a server runs stale code after an upstream fix.
FinOps snapshot — org billing, workflow stats, cache usage. Use when you want a high-level view of CI spending or workflow health before diving deeper.
GitHub Actions billing, workflow efficiency, and waste analysis at org or repo level. Use when investigating CI/CD costs, wasted runs, or optimizing triggers.
正在显示 SKILL.md
基于 SOC 职业分类
| created | "2025-12-16T00:00:00.000Z" |
| modified | "2026-05-09T00:00:00.000Z" |
| reviewed | "2025-12-16T00:00:00.000Z" |
| allowed-tools | Read, Write, Edit, MultiEdit, Bash(git *), Bash(pytest *), Bash(npm test *), Bash(cargo test *), Bash(go test *), mcp__github__list_issues, mcp__github__create_issue, mcp__github__create_pull_request, mcp__github__get_issue, TodoWrite |
| args | [--max-cycles <n>] [--focus <bug|feature|test>] |
| argument-hint | [--max-cycles <n>] [--focus <bug|feature|test>] |
| disable-model-invocation | true |
| description | Automated dev loop — run tests, file issues for failures, TDD fix on branch, open PR, watch CI. Use when running a continuous dev loop, autonomous TDD, or fix-and-PR cycles. |
| name | workflow-dev |
Automated development loop with issue creation, TDD, and CI monitoring.
| Use this skill when... | Use X instead when... |
|---|---|
| Running a continuous loop that turns test failures into issues, then resolves them | Inspecting an existing workflow run or debugging CI failures (/workflow:github-actions-inspection) |
| Automating the full issue → branch → PR → green-CI cycle through the backlog | Generating a reusable auto-fix workflow file for multiple repos (/workflow:github-workflow-auto-fix --reusable) |
| Wanting Claude to pick the next issue and drive it through TDD without user input | Searching upstream OSS issues for known errors or workarounds (/workflow:github-issue-search) |
git branch --show-currentgit status --porcelainfind . -maxdepth 1 \( -name "package.json" -o -name "Cargo.toml" -o -name "pyproject.toml" -o -name "go.mod" -o -name "manage.py" \) -type fOpen issues are fetched during execution (requires a configured git remote).
Parse from $ARGUMENTS:
--max-cycles <n>: Limit to N issue resolution cycles (default: unlimited)--focus <bug|feature|test>: Only work on issues with matching label--quick-wins: Only pick issues estimated < 30 minutes--test-only: Only create issues for test failures; skip implementation--dry-run: Explain what would be done without making changesExecute this automated development loop:
1. Ensure on main branch and sync
git switch main
git pull
2. Run the project's test suite
manage.py, package.json, Cargo.toml, etc.)python manage.py testnpm testcargo testpytest or python -m pytestgo test ./...3. Create GitHub issues for any test failures
github:create_issue with:
"Test failure: {test_name}"["bug", "test-failure"]4. List current repository issues
github:list_issues with state=openSelect issues using this priority order:
priority: critical or bug + high prioritygood first issue or estimated < 1 hourtechnical-debt or refactorSelection criteria (must meet ALL):
blocked or waiting-for-inputIf no suitable issues found:
npm audit, pip-audit, cargo audit)For the selected issue, repeat until CI passes:
5. Create feature branch
git switch -c fix/issue-{number}-{brief-description}
6. Gather implementation context
context7:resolve-library-id followed by context7:get-library-docs for any tools mentioned7. Implement solution using TDD
8. Commit changes
git add <files>
git commit -m "fix: resolve issue #{number} - {brief description}"
9. Push branch
git push origin fix/issue-{number}-{brief-description}
10. Create pull request
github:create_pull_request with:
"fix: resolve issue #{number} - {brief description}""Fixes #{number}" plus description of changes11. Monitor CI status
12. Fix CI failures (if any)
github:get_pull_request_status to check specific failures13. Verify CI success
14. Close the issue
github:update_issue to close with state closed15. Return to Step 1
Handle these command variations:
/workflow:dev - Run continuous loop until stopped/workflow:dev --max-cycles 3 - Limit to 3 issue resolution cycles/workflow:dev --focus bug - Only work on issues labeled "bug"/workflow:dev --quick-wins - Only pick issues estimated < 30 minutes/workflow:dev --test-only - Only create issues for test failures; implementation handled separately/workflow:dev --dry-run - Explain what would be done without making changesIssue too complex to complete:
"This issue requires more detailed analysis - skipping for now"needs-investigationPersistent CI failures (3+ attempts):
"CI Pipeline Issue: {workflow_name} failing"ci-infrastructureNo issues available:
npm update, pip install --upgrade, etc.)Git/GitHub API errors:
After each completed issue, report:
✅ Issue #{number} resolved successfully!
Branch: fix/issue-{number}-{description}
PR: #{pr_number} - {pr_title}
CI Status: ✅ All checks passing
Time: {duration}
Continuing to next issue...
Stop the loop when:
--max-cycles limit reachedProvide final summary:
🏁 DevLoop completed!
📊 Summary:
- Issues resolved: {count}
- PRs created: {count}
- Tests added/fixed: {count}
- Total time: {duration}
📋 Recommendations:
- Review open PRs for merge
- Consider upgrading dependencies
- Add more test coverage for uncovered code
Required tools:
Repository requirements:
Best practices: