{"id":46794411,"url":"https://github.com/mkaczkowski/ai-command-library","last_synced_at":"2026-03-10T03:34:22.271Z","repository":{"id":315520511,"uuid":"1058283432","full_name":"mkaczkowski/ai-command-library","owner":"mkaczkowski","description":null,"archived":false,"fork":false,"pushed_at":"2026-01-31T21:08:31.000Z","size":449,"stargazers_count":2,"open_issues_count":1,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-02-12T16:58:00.150Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/mkaczkowski.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2025-09-16T22:07:13.000Z","updated_at":"2026-01-31T21:08:32.000Z","dependencies_parsed_at":"2025-09-19T05:27:53.114Z","dependency_job_id":"5e5260e8-60df-4503-a958-6b64406eafe0","html_url":"https://github.com/mkaczkowski/ai-command-library","commit_stats":null,"previous_names":["mkaczkowski/ai-command-library"],"tags_count":31,"template":false,"template_full_name":null,"purl":"pkg:github/mkaczkowski/ai-command-library","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mkaczkowski%2Fai-command-library","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mkaczkowski%2Fai-command-library/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mkaczkowski%2Fai-command-library/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mkaczkowski%2Fai-command-library/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/mkaczkowski","download_url":"https://codeload.github.com/mkaczkowski/ai-command-library/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mkaczkowski%2Fai-command-library/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":30323230,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-03-10T01:36:58.598Z","status":"online","status_checked_at":"2026-03-10T02:00:06.579Z","response_time":106,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-03-10T03:34:21.574Z","updated_at":"2026-03-10T03:34:22.250Z","avatar_url":"https://github.com/mkaczkowski.png","language":"JavaScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🚀 AI Command Library\n\n**Stop writing the same AI prompts over and over.**\n\nGet proven commands for code reviews, PR management, and development workflows that work across Claude, Cursor, Copilot, and Codex.\n\n```bash\n# Install once, works everywhere\nnpx @mkaczkowski/ai-command-library -p claude\n```\n\n**✅ One install → Works everywhere**\n**⚡ Ready-to-use workflows**\n**🔄 Auto-sync across tools**\n**📚 Battle-tested commands**\n\n---\n\n## Why This Matters\n\nEvery developer wastes time on:\n\n- 🔄 Writing similar prompts for code reviews and PR management\n- 🎯 Getting inconsistent AI results across the team\n- ⏰ Recreating workflows across projects\n- 🤝 Unable to share effective AI commands\n\n**This library fixes all of that.**\n\nYou get pre-built workflows, consistent setup, team collaboration, and instant productivity.\n\n---\n\n## Quick Start\n\nGet your first AI command working in 2 minutes:\n\n```bash\n# For Claude Desktop/Code\nnpx @mkaczkowski/ai-command-library -p claude\n\n# For Cursor IDE\nnpx @mkaczkowski/ai-command-library -p cursor\n\n# For GitHub Copilot\nnpx @mkaczkowski/ai-command-library -p copilot\n\n# For Codex CLI\nnpx @mkaczkowski/ai-command-library -p codex-global\n```\n\n**Try it now:**\n\n\u003e \"Use the PR review enhancement workflow to improve my code review comments\"\n\n🎉 Done! Your AI tools now have professional workflows.\n\n---\n\n## How It Works\n\nThink of this as **shared AI prompts organized into plugins**. Instead of writing your own, you get:\n\n```\nplugins/pr/              → PR workflows + scripts\nplugins/debugger/        → Debugging tools\nplugins/jira/            → JIRA integration\n       ↓\n  link-ai-commands       → Smart sync\n       ↓\n.claude/commands/        → Commands\n.claude/skills/          → Skills (Claude only)\n.claude/agents/          → Agents (Claude only)\n```\n\n**Core concepts:**\n\n- 📦 **Plugins** - Grouped commands for specific tasks (PR, debugging, JIRA)\n- 📝 **Commands** - Step-by-step AI instructions\n- 💡 **Skills** - Reusable capabilities (Claude only)\n- 🎯 **Agents** - Specialized subagents (Claude only)\n- 🔗 **Linking** - Auto-sync to your AI tools\n\n---\n\n## Installation Options\n\n### Direct Usage (No Install)\n\n```bash\nnpx @mkaczkowski/ai-command-library -p \u003cprovider\u003e\n```\n\n### Install \u0026 Link\n\n```bash\nnpm install -D @mkaczkowski/ai-command-library\nnpx link-ai-commands --provider \u003cprovider\u003e\n```\n\n### Auto-Sync on Install\n\n```json\n{\n  \"scripts\": {\n    \"postinstall\": \"link-ai-commands --provider claude\"\n  }\n}\n```\n\n---\n\n## Choose Your Provider\n\n| Provider        | Destination         | Skills | Agents | Best For                |\n| --------------- | ------------------- | ------ | ------ | ----------------------- |\n| `claude`        | `.claude/`          | ✅     | ✅     | Project-specific Claude |\n| `claude-global` | `~/.claude/`        | ✅     | ✅     | User-wide Claude        |\n| `cursor`        | `.cursor/`          | ❌     | ❌     | Cursor IDE              |\n| `cursor-global` | `~/.cursor/`        | ❌     | ❌     | User-wide Cursor        |\n| `copilot`       | `.github/prompts/`  | ❌     | ❌     | Copilot Workspace       |\n| `codex-global`  | `~/.codex/prompts/` | ❌     | ❌     | Codex CLI               |\n\n**Provider-specific notes:**\n\n- **Copilot:** Prompts are flattened to hyphenated names ending in `.prompt.md`\n- **Codex:** Commands are flattened to single filenames (e.g., `pr__enhance-review.md`)\n- **Skills/Agents:** Only available for Claude providers (project and global)\n\n**Multi-tool setup:**\n\n```bash\n# Use multiple AI tools? Run once per provider\nnpx link-ai-commands --provider claude\nnpx link-ai-commands --provider cursor\n```\n\n---\n\n## Advanced Options\n\n**Install specific plugins:**\n\n```bash\nnpx link-ai-commands --provider claude --plugins pr,debugger\n```\n\n**List available plugins:**\n\n```bash\nnpx link-ai-commands --list-groups\n```\n\n**Use symlinks for live updates:**\n\n```bash\nnpx link-ai-commands --provider claude --mode symlink\n```\n\n**Preview changes:**\n\n```bash\nnpx link-ai-commands --provider claude --dry-run\n```\n\n---\n\n## Available Workflows\n\nReady-to-use workflows that solve real problems. Each includes commands + helper scripts for GitHub integration.\n\n### 🔧 PR Review Enhancement\n\nTransform blunt code review feedback into collaborative, professional guidance.\n\n**Command:** `pr/enhance-review.md`\n\n**When to use:** After writing initial review comments but before posting them.\n\n```bash\n# Step 1: Fetch and rewrite comments\nnode {{script:pr/scripts/fetch-pr-comments.js}} --pr=123\n# Then ask AI: \"Enhance my review comments\"\n# Output: tmp/pr-[PR_NUMBER]-comments.md\n\n# Step 2: Update GitHub\nnode {{script:pr/scripts/edit-pr-comments.js}} --pr=123\n```\n\n**Benefits:** Professional tone, same technical accuracy, stronger team relationships.\n\n\u003cdetails\u003e\n\u003csummary\u003eView workflow diagram\u003c/summary\u003e\n\n```mermaid\nflowchart TD\n    Fetch[Fetch PR comments]\n    Fetch --\u003e Context[Review diff and project standards]\n    Context --\u003e Rewrite[Rewrite in collaborative voice]\n    Rewrite --\u003e Output[Save improved feedback]\n    Output --\u003e Decide{Push to GitHub?}\n    Decide --\u003e|Yes| Update[Update comments]\n    Decide --\u003e|No| Share[Manual review]\n```\n\n\u003c/details\u003e\n\n---\n\n### ✍️ Create PR Reviews\n\nAI-powered systematic code review with comprehensive issue detection.\n\n**Command:** `pr/draft-review.md`\n\n**When to use:** When you need to review a new PR thoroughly.\n\n```bash\n# Step 1: Prepare review findings\nnode {{script:pr/scripts/fetch-pr-context.js}} --pr=123\n# Then ask AI: \"Prepare a review for this PR\"\n# Output: tmp/pr-[PR_NUMBER]-findings.md\n\n# Step 2: Create and post review\n# AI generates: tmp/pr-[PR_NUMBER]-review-comments.json\nnode {{script:pr/scripts/create-pr-review.js}} --input=tmp/pr-123-review-comments.json --pr=123\n```\n\n**Benefits:** Catch issues you might miss, professional comments, faster reviews.\n\n\u003cdetails\u003e\n\u003csummary\u003eView workflow diagram\u003c/summary\u003e\n\n```mermaid\nflowchart TD\n    Context[Fetch PR context]\n    Context --\u003e Inspect[Study patches and neighboring code]\n    Inspect --\u003e Analyze[Assess risks and strengths]\n    Analyze --\u003e Output[Generate structured findings]\n    Output --\u003e Craft[Compose inline comments]\n    Craft --\u003e Map[Align paths and line numbers]\n    Map --\u003e Export[Export to JSON]\n    Export --\u003e RunScript[Create pending PR review]\n```\n\n\u003c/details\u003e\n\n---\n\n### ✨ Create Pull Requests\n\nAuto-generate comprehensive PR descriptions from your code changes.\n\n**Command:** `pr/create-pr.md`\n\n**When to use:** When you're ready to create a pull request from your feature branch.\n\n```bash\n# AI analyzes your changes and populates the PR template\n# Ask AI: \"Create a PR using the create-pr workflow\"\n\n# Or provide context upfront:\n# \"Create a PR. I added a user search feature with debouncing to reduce API calls.\"\n```\n\n**What the AI does:**\n\n1. Reads `.github/PULL_REQUEST_TEMPLATE.md`\n2. Analyzes commits and diffs since diverging from main\n3. Extracts tickets from branch name/commits\n4. Identifies new dependencies and test coverage\n5. Populates template with reviewer-friendly content\n6. Creates PR using `gh pr create`\n\n**Benefits:** Complete descriptions, faster reviews, consistent quality.\n\n---\n\n### 🔗 Create Stacked Pull Requests\n\n**Problem:** Large features are hard to review in a single PR, and breaking them into dependent PRs manually is time-consuming and error-prone.\n\n**Solution:** Automated creation of sequential dependent PRs that build on each other, following conventional commit format with proper dependency tracking.\n\n**What you get:**\n\n- 📚 Work broken into logical, reviewable chunks\n- 🔗 Proper dependency chain between PRs\n- 🎯 Conventional commit titles with stack position tracking\n- ⚡ Faster reviews through smaller, focused PRs\n\n#### Create Stacked PRs (`pr/create-stacked-pr.md`)\n\n**When to use:** When you have a large feature that should be broken into multiple dependent PRs for easier review.\n\n```bash\n# Ask your AI:\n# \"Create stacked PRs for feat auth PROJ-123 implement OAuth2 authentication flow\"\n```\n\n**What happens:** Your AI:\n\n1. Analyzes the full scope of work needed\n2. Breaks it into logical chunks (foundation → features)\n3. Plans the stack with proper dependencies\n4. Shows you the plan for approval\n5. Creates sequential branches with conventional commit format\n6. Generates dependent PRs with proper base branches\n7. Provides a summary with all PR links and merge order\n\n**What you need to provide:**\n\n- Conventional commit type (feat, fix, refactor, etc.)\n- Scope/feature area\n- Ticket number\n- Description of the changes\n\n**Output:** Series of dependent PRs with proper titles like `feat(auth): add token validation 1/3 (PROJ-123)`\n\n### 🔄 Address Review Feedback\n\nSystematically resolve all review comments with tracking and validation.\n\n**Command:** `pr/address-review.md`\n\n**When to use:** When you have unresolved review comments to address.\n\n```bash\n# Step 1: Plan resolutions\nnode {{script:pr/scripts/fetch-pr-comments.js}} --pr=123 --reaction=+1 --ignore-outdated --include-diff-hunk\n# Then ask AI: \"Plan resolutions for this feedback\"\n# Output: tmp/pr-[PR_NUMBER]-address-plan.md\n\n# Step 2: Implement changes\n# AI guides you through each planned change\n# Output: tmp/pr-[PR_NUMBER]-address-report.md\n\n# Step 3: Reply to resolved comments\nnode {{script:pr/scripts/reply-to-comments.js}} --pr=123 --dry-run\nnode {{script:pr/scripts/reply-to-comments.js}} --pr=123\n```\n\n**Benefits:** Nothing gets missed, appropriate responses, faster resolution cycles.\n\n\u003cdetails\u003e\n\u003csummary\u003eView workflow diagram\u003c/summary\u003e\n\n```mermaid\nflowchart TD\n    Gather[Fetch unresolved threads + PR context]\n    Gather --\u003e Study[Review diffs and previous discussion]\n    Study --\u003e Strategize[Define fixes and validation per comment]\n    Strategize --\u003e Document[Write implementation plan]\n    Document --\u003e Implement[Apply fixes, docs, and tests]\n    Implement --\u003e Validate[Run targeted checks]\n    Validate --\u003e Report[Generate final report]\n    Report --\u003e Reply[Reply with commit references]\n    Reply --\u003e Wrap{Everything resolved?}\n    Wrap --\u003e|Yes| Share[Push commits and close loop]\n    Wrap --\u003e|No| Loop[Return to remaining feedback]\n```\n\n\u003c/details\u003e\n\n---\n\n### 🐞 File JIRA Tickets\n\nQuick JIRA ticket creation with guided workflows:\n\n#### Bug Reports (`jira/create-bug.template.md`)\n\n```bash\n# Ask AI: \"File a JIRA bug for [issue]\"\n```\n\nOutputs: Complete description + ready-to-run `jira issue create` command\n\n#### Technical Debt (`jira/create-tech-debt.template.md`)\n\n```bash\n# Ask AI: \"Create a technical debt ticket for [work]\"\n```\n\nOutputs: Structured ticket with What/Where/Why + acceptance criteria\n\n#### Flaky Tests (`jira/create-flaky-test.template.md`)\n\n```bash\n# Ask AI: \"File a flaky test ticket for [test]\"\n```\n\nOutputs: Complete diagnostic details + reproduction steps\n\n**Prerequisite:** Install [`jira-cli`](https://github.com/ankitpokhrel/jira-cli) and authenticate.\n\n---\n\n## Script Integration\n\nCommands reference helper scripts using placeholders that work everywhere:\n\n```bash\n# This works across all providers:\nnode {{script:pr/scripts/fetch-pr-comments.js}} --pr=123\n\n# Resolves to correct path for each provider:\n# Claude:  .claude/commands/pr/scripts/...\n# Cursor:  .cursor/commands/pr/scripts/...\n# Copilot: .github/prompts/pr/scripts/...\n# Codex:   ~/.codex/prompts/pr/scripts/...\n```\n\n**Available placeholders:**\n\n- `{{script:path/to/script.js}}` - References helper scripts\n- `{{path:commandsRoot}}` - References commands directory\n\n---\n\n## Helper Scripts Reference\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003e📥 Data Fetching\u003c/strong\u003e\u003c/summary\u003e\n\n- `fetch-pr-comments.js` - Get review comments with filtering\n- `fetch-pr-context.js` - Gather PR metadata, files, commits\n\n```bash\nnode {{script:pr/scripts/fetch-pr-comments.js}} --pr=123 --reaction=+1\nnode {{script:pr/scripts/fetch-pr-context.js}} --pr=123\n```\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003e📤 GitHub Integration\u003c/strong\u003e\u003c/summary\u003e\n\n- `create-pr-review.js` - Submit comment JSON as review\n- `edit-pr-comments.js` - Update existing comments\n- `reply-to-comments.js` - Post resolution acknowledgments\n\n```bash\nnode {{script:pr/scripts/create-pr-review.js}} --input=tmp/pr-123-review.json --pr=123\nnode {{script:pr/scripts/edit-pr-comments.js}} --pr=123\nnode {{script:pr/scripts/reply-to-comments.js}} --pr=123 --dry-run\n```\n\n\u003c/details\u003e\n\n💡 All scripts support `--help` for usage details.\n\n---\n\n## Prerequisites\n\n- **Node.js 22+** and **npm 10+**\n- **[GitHub CLI](https://cli.github.com/)** for PR workflows\n  - Run `gh auth login` with `repo` scope\n  - Enterprise users: set `GH_HOST` or use `--repo` flags\n\n### Alternative Installation\n\nInstall directly from GitHub:\n\n```bash\nnpm install -D git+https://github.com/mkaczkowski/ai-command-library.git\n```\n\n---\n\n## Development \u0026 Contributing\n\n### Setup\n\n```bash\ngit clone https://github.com/mkaczkowski/ai-command-library.git\ncd ai-command-library\nnpm install\n```\n\n### Code Quality\n\n```bash\nnpm run lint           # Check code style\nnpm run format:check   # Verify formatting\nnpm run lint:fix       # Fix issues\nnpm run format         # Apply formatting\n```\n\n### Repository Structure\n\n```\nai-command-library/\n├── plugins/              # Plugin groups\n│   ├── pr/              # PR workflows\n│   ├── debugger/        # Debugging tools\n│   ├── jira/            # JIRA integration\n│   └── common/          # Shared utilities\n├── providers/           # Provider configs\n├── scripts/             # Core linking logic\n└── bin/                 # CLI entry point\n```\n\n### Plugin Organization\n\nEach plugin group can contain:\n\n```\n{group}/\n├── commands/      # Markdown command templates + scripts\n├── skills/        # Claude Code Skills (Claude only)\n└── agents/        # Claude Code Subagents (Claude only)\n```\n\n**Available groups:**\n\n- **pr/** - Pull request workflows and utilities\n- **debugger/** - Debugging agents and skills\n- **jira/** - JIRA ticket creation\n- **common/** - Cross-cutting utilities\n\n### Release Process\n\n1. Update `package.json` version and `CHANGELOG.md`\n2. Run `npm run release` to verify\n3. Publish: `npm publish --access public`\n4. Push release commit and tag\n\n---\n\n## Support \u0026 Community\n\n### Need Help?\n\n- 🐛 **Bug reports:** [GitHub Issues](https://github.com/mkaczkowski/ai-command-library/issues)\n- 💡 **Feature requests:** [GitHub Issues](https://github.com/mkaczkowski/ai-command-library/issues)\n- 🔒 **Security concerns:** See [`SECURITY.md`](SECURITY.md)\n\n### Contributing\n\nWe welcome contributions:\n\n- Add new workflows\n- Improve existing templates\n- Enhance provider support\n- Update documentation\n\n---\n\n## License\n\nReleased under the [MIT License](LICENSE).\n\n**Happy coding!** 🚀\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmkaczkowski%2Fai-command-library","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmkaczkowski%2Fai-command-library","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmkaczkowski%2Fai-command-library/lists"}