[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-74243":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":15,"stars7d":16,"stars30d":13,"stars90d":15,"forks30d":15,"starsTrendScore":15,"compositeScore":17,"rankGlobal":9,"rankLanguage":9,"license":18,"archived":19,"fork":19,"defaultBranch":20,"hasWiki":21,"hasPages":19,"topics":22,"createdAt":9,"pushedAt":9,"updatedAt":31,"readmeContent":32,"aiSummary":33,"trendingCount":15,"starSnapshotCount":15,"syncStatus":34,"lastSyncTime":35,"discoverSource":36},74243,"claude-workflow-v2","CloudAI-X\u002Fclaude-workflow-v2","CloudAI-X","Universal Claude Code workflow plugin with agents, skills, hooks, and commands",null,"Python",1366,191,13,5,0,4,19.85,"MIT License",false,"main",true,[23,24,25,26,27,28,29,30],"agent-skills","ai","ai-agents","claude-code","codex","cursor","skills","workflow","2026-06-12 02:03:24","# project-starter\n\n[![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-yellow.svg)](https:\u002F\u002Fopensource.org\u002Flicenses\u002FMIT)\n[![Claude Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FClaude%20Code-v1.0.33+-blue.svg)](https:\u002F\u002Fcode.claude.com)\n[![PRs Welcome](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPRs-welcome-brightgreen.svg)](https:\u002F\u002Fgithub.com\u002FCloudAI-X\u002Fclaude-workflow-v2\u002Fpulls)\n\nA universal Claude Code workflow plugin with specialized agents, skills, hooks, and output styles for any software project. Compatible with [skills.sh](https:\u002F\u002Fskills.sh) — works with Claude Code, Cursor, Codex, and 35+ AI agents.\n\n---\n\n## Quick Start\n\n### Option 1: skills.sh (Recommended — Any Agent)\n\n```bash\nnpx skills add CloudAI-X\u002Fclaude-workflow-v2\n```\n\nInstalls skills to Claude Code, Cursor, Codex, Windsurf, Cline, and 35+ other AI agents automatically.\n\n### Option 2: npx (Claude Code — Full Plugin)\n\n```bash\nnpx install-claude-workflow-v2@latest\n```\n\nInstalls the complete plugin: agents, commands, skills, and hooks.\n\n### Option 3: CLI (Per-Session)\n\n```bash\n# Clone the plugin\ngit clone https:\u002F\u002Fgithub.com\u002FCloudAI-X\u002Fclaude-workflow-v2.git\n\n# Run Claude Code with the plugin\nclaude --plugin-dir .\u002Fclaude-workflow-v2\n```\n\n### Option 4: Agent SDK\n\n```typescript\nimport { query } from \"@anthropic-ai\u002Fclaude-agent-sdk\";\n\nfor await (const message of query({\n  prompt: \"Hello\",\n  options: {\n    plugins: [{ type: \"local\", path: \".\u002Fclaude-workflow-v2\" }],\n  },\n})) {\n  \u002F\u002F Plugin commands, agents, and skills are now available\n}\n```\n\n### Option 5: Install Permanently\n\n```bash\n# Install from marketplace (when available)\nclaude plugin install project-starter\n\n# Or install from local directory\nclaude plugin install .\u002Fclaude-workflow-v2\n```\n\n### Verify Installation\n\nAfter loading the plugin, verify it's working:\n\n```\n> \u002Fplugin\n```\n\nTab to **Installed** - you should see `project-starter` listed.\nTab to **Errors** - should be empty (no errors).\n\nThese commands become available:\n\n```\n\u002Fproject-starter:architect    # Architecture-first mode\n\u002Fproject-starter:rapid        # Ship fast mode\n\u002Fproject-starter:commit       # Auto-generate commit message\n\u002Fproject-starter:verify-changes  # Multi-agent verification\n```\n\n---\n\n## What's Included\n\n| Component    | Count | Description                                                             |\n| ------------ | ----- | ----------------------------------------------------------------------- |\n| **Agents**   | 7     | Specialized subagents for code review, debugging, security, etc.        |\n| **Commands** | 26    | Slash commands for workflows, output styles, planning, and onboarding   |\n| **Skills**   | 14    | Knowledge domains with on-demand context loading                        |\n| **Hooks**    | 14    | Automation scripts for formatting, security, metrics, and notifications |\n\n---\n\n## Usage Examples\n\n### Commands in Action\n\n**Auto-commit your changes:**\n\n```\n> \u002Fproject-starter:commit\n\nLooking at staged changes...\n✓ Created commit: feat(auth): add JWT refresh token endpoint\n```\n\n**Full git workflow:**\n\n```\n> \u002Fproject-starter:commit-push-pr\n\n✓ Committed: feat: add user dashboard\n✓ Pushed to origin\u002Ffeature\u002Fdashboard\n✓ Created PR #42: https:\u002F\u002Fgithub.com\u002Fyou\u002Frepo\u002Fpull\u002F42\n```\n\n**Verify before shipping:**\n\n```\n> \u002Fproject-starter:verify-changes\n\nSpawning verification agents...\n├─ build-validator: ✓ Build passes\n├─ test-runner: ✓ 42 tests pass\n├─ lint-checker: ⚠ 2 warnings (non-blocking)\n└─ security-scanner: ✓ No vulnerabilities\n\nReady to ship!\n```\n\n### Agents in Action\n\nAgents spawn automatically based on your request:\n\n**You say:** \"The login is broken, users get 401 errors\"\n\n```\n[debugger agent activated]\n→ Checking auth middleware... found issue\n→ Token validation uses wrong secret in production\n→ Fix: Update AUTH_SECRET in .env.production\n```\n\n**You say:** \"Review my changes\"\n\n```\n[code-reviewer agent activated]\n→ Analyzing 3 files changed...\n✓ Logic is correct\n⚠ Missing null check on line 42\n⚠ Consider adding rate limiting to this endpoint\n```\n\n**You say:** \"Add authentication to the API\"\n\n```\n[orchestrator agent activated]\n→ Breaking down into subtasks:\n  1. Design auth schema (spawning architect)\n  2. Implement JWT middleware\n  3. Add login\u002Fregister endpoints\n  4. Write tests (spawning test-architect)\n  5. Update API docs (spawning docs-writer)\n```\n\n### Skills in Action\n\nSkills provide domain knowledge automatically:\n\n**You ask:** \"How should I structure the payment service?\"\n\n```\n[designing-architecture skill applied]\n→ Recommending hexagonal architecture\n→ Payment providers as adapters\n→ Core domain isolated from infrastructure\n```\n\n**You ask:** \"Make this endpoint faster\"\n\n```\n[optimizing-performance skill applied]\n→ Adding database indexes\n→ Implementing response caching\n→ Using pagination for large results\n```\n\n### Hooks in Action\n\nHooks run automatically on events:\n\n**Security block (pre-edit):**\n\n```\n⛔ BLOCKED: Potential secret detected\n   File: src\u002Fconfig.ts, Line 5\n   Pattern: API key (sk-...)\n\n   Remove the secret and use environment variables.\n```\n\n**Auto-format (post-edit):**\n\n```\n✓ Formatted with prettier: src\u002Fcomponents\u002FButton.tsx\n✓ Formatted with black: scripts\u002Fdeploy.py\n```\n\n**Desktop notifications:**\n\n```\n🔔 \"Claude needs input\" - when waiting for your response\n🔔 \"Task complete\" - when finished\n```\n\n---\n\n## Commands Reference\n\nAll commands use the format `\u002Fproject-starter:\u003Ccommand>`.\n\n### Output Styles\n\n| Command                      | Mode                                          |\n| ---------------------------- | --------------------------------------------- |\n| `\u002Fproject-starter:architect` | System design mode - architecture before code |\n| `\u002Fproject-starter:rapid`     | Fast development - ship quickly, iterate      |\n| `\u002Fproject-starter:mentor`    | Teaching mode - explain the \"why\"             |\n| `\u002Fproject-starter:review`    | Code review mode - strict quality             |\n\n### Git Workflow (Inner-Loop)\n\n| Command                              | Purpose                                   |\n| ------------------------------------ | ----------------------------------------- |\n| `\u002Fproject-starter:commit`            | Auto-generate conventional commit message |\n| `\u002Fproject-starter:commit-push-pr`    | Commit → Push → Create PR (full workflow) |\n| `\u002Fproject-starter:quick-fix`         | Fast fix for lint\u002Ftype errors             |\n| `\u002Fproject-starter:add-tests`         | Generate tests for recent changes         |\n| `\u002Fproject-starter:lint-fix`          | Auto-fix all linting issues               |\n| `\u002Fproject-starter:sync-branch`       | Sync with main (rebase or merge)          |\n| `\u002Fproject-starter:summarize-changes` | Generate standup\u002FPR summaries             |\n\n### Verification\n\n| Command                            | Purpose                                 |\n| ---------------------------------- | --------------------------------------- |\n| `\u002Fproject-starter:verify-changes`  | Multi-subagent adversarial verification |\n| `\u002Fproject-starter:validate-build`  | Build process validation                |\n| `\u002Fproject-starter:run-tests`       | Tiered test execution                   |\n| `\u002Fproject-starter:lint-check`      | Code quality checks                     |\n| `\u002Fproject-starter:security-scan`   | Security vulnerability detection        |\n| `\u002Fproject-starter:code-simplifier` | Post-implementation cleanup             |\n\n### Planning & Refactoring\n\n| Command                               | Purpose                                    |\n| ------------------------------------- | ------------------------------------------ |\n| `\u002Fproject-starter:plan`               | Persistent PLAN.md with phase tracking     |\n| `\u002Fproject-starter:refactor-guided`    | 4-phase systematic refactoring with safety |\n| `\u002Fproject-starter:dependency-upgrade` | Safe dependency upgrades with rollback     |\n\n### Onboarding & Knowledge\n\n| Command                                   | Purpose                                   |\n| ----------------------------------------- | ----------------------------------------- |\n| `\u002Fproject-starter:tutorial`               | Interactive guided tutorial for new users |\n| `\u002Fproject-starter:bootstrap-repo`         | 10-agent parallel repo exploration        |\n| `\u002Fproject-starter:save-session-learnings` | Persist session discoveries to docs       |\n| `\u002Fproject-starter:metrics`                | View agent performance metrics            |\n\n---\n\n## Agents\n\nAgents are specialized subagents that Claude spawns automatically based on your task.\n\n| Agent              | Purpose                          | Auto-Triggers                                                |\n| ------------------ | -------------------------------- | ------------------------------------------------------------ |\n| `orchestrator`     | Coordinate multi-step tasks      | \"improve\", \"enhance\", \"build\", \"architecture\", complex tasks |\n| `code-reviewer`    | Review code quality              | \"review\", \"PR review\", \"lint\", code changes                  |\n| `debugger`         | Systematic bug investigation     | Errors, crashes, memory leaks, timeouts, race conditions     |\n| `docs-writer`      | Technical documentation          | README, changelogs, migration guides, release notes          |\n| `security-auditor` | Security vulnerability detection | Auth, encryption, secrets, OAuth, JWT, CORS                  |\n| `refactorer`       | Code structure improvements      | Tech debt, code smells, complexity reduction                 |\n| `test-architect`   | Design test strategies           | Test plans, mocking, flaky tests, integration\u002FE2E            |\n\n---\n\n## Skills\n\nSkills are knowledge domains that Claude uses autonomously when relevant.\n\n| Skill                         | Domain                                                |\n| ----------------------------- | ----------------------------------------------------- |\n| `analyzing-projects`          | Understand codebase structure and patterns            |\n| `designing-tests`             | Unit, integration, E2E test approaches                |\n| `designing-architecture`      | Clean Architecture, Hexagonal, etc.                   |\n| `optimizing-performance`      | Speed up applications, identify bottlenecks           |\n| `managing-git`                | Version control, conventional commits                 |\n| `designing-apis`              | REST\u002FGraphQL patterns and best practices              |\n| `parallel-execution`          | Multi-subagent parallel task execution patterns       |\n| `web-design-guidelines`       | Self-contained UI audit (A11Y, PERF, RD, SEC, I18N)   |\n| `vercel-react-best-practices` | React\u002FNext.js performance optimization (45 rules)     |\n| `convex-backend`              | Convex backend development (functions, schemas, etc.) |\n| `database-design`             | Schema design, indexing, query optimization           |\n| `devops-infrastructure`       | Docker, CI\u002FCD, deployment, IaC, monitoring            |\n| `error-handling`              | Error patterns, structured logging, retry\u002Fcircuit     |\n| `security-patterns`           | Auth, RBAC, secrets, CORS, rate limiting, headers     |\n\n---\n\n## Hooks\n\nHooks run automatically on specific events.\n\n| Hook                  | Trigger       | Action                                  |\n| --------------------- | ------------- | --------------------------------------- |\n| Security scan         | Edit\u002FWrite    | Blocks commits with potential secrets   |\n| File protection       | Edit\u002FWrite    | Blocks edits to lock files, .env, .git  |\n| Auto-format           | Edit\u002FWrite    | Runs prettier\u002Fblack\u002Fgofmt by file type  |\n| TypeScript check      | Edit\u002FWrite    | Runs `tsc --noEmit` on .ts\u002F.tsx files   |\n| Pre-commit check      | Bash          | Detects debug statements & temp markers |\n| Branch protection     | Bash          | Warns on commits to protected branches  |\n| Command logging       | Bash          | Logs to `.claude\u002Fcommand-history.log`   |\n| Environment check     | Session start | Validates Node.js, Python, Git          |\n| Prompt analysis       | User prompt   | Suggests appropriate agents             |\n| Auto-verify           | Task complete | Runs tests\u002Flint, reports results        |\n| Doc update suggest    | Task complete | Suggests CLAUDE.md updates for changes  |\n| Session metrics       | Task complete | Logs session telemetry to metrics file  |\n| Input notification    | Input needed  | Desktop notification                    |\n| Complete notification | Task complete | Desktop notification                    |\n\n---\n\n## Examples\n\nFor detailed multi-agent orchestration examples, see the [examples\u002F](.\u002Fexamples\u002F) directory:\n\n| Example                                                                          | Description                                            |\n| -------------------------------------------------------------------------------- | ------------------------------------------------------ |\n| [Comprehensive Code Review](.\u002Fexamples\u002Forchestration\u002Fcomprehensive-code-review\u002F) | 6-agent sequential workflow for thorough code analysis |\n\nEach example includes:\n\n- **README.md** - Overview and quick start\n- **workflow.md** - Exact prompts to use\n- **verification.md** - How to verify it works\n- **sample-outputs\u002F** - Example agent outputs\n\n---\n\n## Configuration\n\n### Add Permissions to Your Project\n\nCopy the permissions template to your project:\n\n```bash\nmkdir -p \u002Fpath\u002Fto\u002Fyour\u002Fproject\u002F.claude\ncp templates\u002Fsettings.local.json.template \u002Fpath\u002Fto\u002Fyour\u002Fproject\u002F.claude\u002Fsettings.local.json\n```\n\nThis pre-allows common safe commands so you don't get prompted every time.\n\n### Add Team Conventions\n\nCopy the CLAUDE.md template to your project root:\n\n```bash\ncp templates\u002FCLAUDE.md.template \u002Fpath\u002Fto\u002Fyour\u002Fproject\u002FCLAUDE.md\n```\n\nThen customize with your:\n\n- Package manager commands\n- Test\u002Fbuild\u002Flint commands\n- Code conventions\n- Architecture decisions\n\n### MCP Servers\n\nCopy the MCP template to enable integrations like Slack, GitHub, Sentry:\n\n```bash\ncp templates\u002Fmcp.json.template \u002Fpath\u002Fto\u002Fyour\u002Fproject\u002F.mcp.json\n```\n\nThen configure the environment variables for the servers you want to use.\n\n### GitHub Action (@.claude in PRs)\n\nEnable Claude to respond to PR comments by installing the GitHub Action:\n\n```bash\n# In your repository\nclaude \u002Finstall-github-action\n```\n\nThis enables:\n\n- Tag `@claude` in PR comments to get code suggestions\n- Auto-update `CLAUDE.md` during code review\n- Claude responds to review feedback automatically\n\n**Example PR comment:**\n\n```\n@claude please add input validation to the email field\n```\n\n**Team workflow tip:** Use `@claude` to update your `CLAUDE.md` with learnings from code review:\n\n```\n@claude add a note to CLAUDE.md that we should always validate email format before API calls\n```\n\n---\n\n## Extending the Plugin\n\n### Add Custom Commands\n\nCreate `.md` files in `commands\u002F`:\n\n```markdown\n---\nallowed-tools: Bash(git:*), Read, Write\ndescription: What this command does\nargument-hint: [optional arguments]\n---\n\n[Command instructions here]\n```\n\n### Add Custom Agents\n\nCreate `.md` files in `agents\u002F`:\n\n```markdown\n---\nname: my-agent\ndescription: What it does. Use PROACTIVELY when [triggers].\ntools: Read, Write, Edit, Bash\nmodel: sonnet\n---\n\n[Agent instructions here]\n```\n\n### Add Custom Skills\n\nCreate subdirectories in `skills\u002F` with a `SKILL.md` file:\n\n```markdown\n---\nname: my-skill\ndescription: Guides [domain]. Use when [triggers].\n---\n\n[Skill knowledge and patterns here]\n```\n\n---\n\n## Plugin Structure\n\n```\nclaude-workflow\u002F\n├── .claude-plugin\u002F\n│   ├── plugin.json           # Required: Plugin manifest\n│   └── marketplace.json      # Optional: Marketplace metadata\n├── agents\u002F                   # 7 specialized agents\n│   ├── orchestrator.md\n│   ├── code-reviewer.md\n│   ├── debugger.md\n│   ├── docs-writer.md\n│   ├── security-auditor.md\n│   ├── refactorer.md\n│   └── test-architect.md\n├── commands\u002F                 # 26 slash commands\n│   ├── architect.md          # Output styles\n│   ├── rapid.md\n│   ├── mentor.md\n│   ├── review.md\n│   ├── commit.md             # Git workflow\n│   ├── commit-push-pr.md\n│   ├── quick-fix.md\n│   ├── add-tests.md\n│   ├── lint-fix.md\n│   ├── sync-branch.md\n│   ├── summarize-changes.md\n│   ├── verify-changes.md     # Verification\n│   ├── validate-build.md\n│   ├── run-tests.md\n│   ├── lint-check.md\n│   ├── security-scan.md\n│   ├── code-simplifier.md\n│   ├── parallel-review.md    # Parallel\n│   ├── parallel-analyze.md\n│   ├── plan.md               # Planning & refactoring\n│   ├── refactor-guided.md\n│   ├── dependency-upgrade.md\n│   ├── tutorial.md           # Onboarding & knowledge\n│   ├── bootstrap-repo.md\n│   ├── save-session-learnings.md\n│   └── metrics.md\n├── skills\u002F                   # 14 knowledge domains\n│   ├── analyzing-projects\u002F\n│   ├── convex-backend\u002F\n│   ├── database-design\u002F\n│   ├── designing-apis\u002F\n│   ├── designing-architecture\u002F\n│   ├── designing-tests\u002F\n│   ├── devops-infrastructure\u002F\n│   ├── error-handling\u002F\n│   ├── managing-git\u002F\n│   ├── optimizing-performance\u002F\n│   ├── parallel-execution\u002F\n│   ├── security-patterns\u002F\n│   ├── vercel-react-best-practices\u002F\n│   └── web-design-guidelines\u002F\n├── hooks\u002F\n│   ├── hooks.json            # Hook configuration\n│   └── 14 automation scripts # Pre\u002Fpost tool, session, metrics, notifications\n├── templates\u002F                # User-copyable templates\n│   ├── CLAUDE.md.template\n│   ├── settings.json.template\n│   ├── settings.local.json.template\n│   └── mcp.json.template\n├── CLAUDE.md                 # Plugin development guidelines\n└── README.md\n```\n\n---\n\n## Requirements\n\n- **Claude Code** v1.0.33 or later\n- **Python 3** (for hook scripts)\n- **Node.js** (optional, for npm commands)\n- **Git** (for version control features)\n\n---\n\n## Multi-Agent Compatibility (skills.sh)\n\nThis repo is fully compatible with [skills.sh](https:\u002F\u002Fskills.sh) — the universal agent skills platform. Our 14 skills work with **38+ AI coding agents**:\n\n| Agent           | Install Method                                                       |\n| --------------- | -------------------------------------------------------------------- |\n| **Claude Code** | `npx skills add CloudAI-X\u002Fclaude-workflow-v2` or full plugin install |\n| **Cursor**      | `npx skills add CloudAI-X\u002Fclaude-workflow-v2`                        |\n| **Codex**       | `npx skills add CloudAI-X\u002Fclaude-workflow-v2`                        |\n| **Windsurf**    | `npx skills add CloudAI-X\u002Fclaude-workflow-v2`                        |\n| **Cline**       | `npx skills add CloudAI-X\u002Fclaude-workflow-v2`                        |\n| **35+ more**    | `npx skills add CloudAI-X\u002Fclaude-workflow-v2`                        |\n\n> **Note:** `npx skills add` installs **skills only**. For the full Claude Code experience (agents, commands, hooks), use `npx install-claude-workflow-v2@latest`.\n\n---\n\n## Contributing\n\nContributions welcome! See [CONTRIBUTING.md](.\u002FCONTRIBUTING.md).\n\n1. Fork the repository\n2. Create a feature branch (`git checkout -b feature\u002Famazing-feature`)\n3. Commit your changes (`git commit -m 'feat: add amazing feature'`)\n4. Push to the branch (`git push origin feature\u002Famazing-feature`)\n5. Open a Pull Request\n\n---\n\n## Star History\n\n[![Star History Chart](https:\u002F\u002Fapi.star-history.com\u002Fsvg?repos=CloudAI-X\u002Fclaude-workflow-v2&type=date&legend=top-left)](https:\u002F\u002Fwww.star-history.com\u002F#CloudAI-X\u002Fclaude-workflow-v2&type=date&legend=top-left)\n\n## Credits\n\n- Plugin created by [@cloudxdev](https:\u002F\u002Fx.com\u002Fcloudxdev)\n- Workflow patterns inspired by [Boris Cherny](https:\u002F\u002Fx.com\u002Fbcherny) (creator of Claude Code)\n\n## License\n\nMIT - see [LICENSE](.\u002FLICENSE) for details.\n","CloudAI-X\u002Fclaude-workflow-v2 是一个通用的 Claude Code 工作流插件，提供了专门的代理、技能、钩子和命令来增强软件项目的开发流程。该项目使用 Python 编写，具有 7 个专为代码审查、调试、安全等任务设计的代理，26 条用于工作流管理、输出样式定制、规划及入职引导的命令，14 项按需加载上下文的知识领域技能，以及 14 个自动化脚本以实现格式化、安全性检查、度量与通知等功能。适用于需要集成 AI 辅助工具提升开发效率的各种场景，如快速原型构建、代码质量保证、持续集成等。",2,"2026-06-11 03:49:39","high_star"]