cmd-example-command — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited cmd-example-command (Rules) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
<div align="center"> <a href="https://github.com/nota-america/forgecat-agent-profiles"> <img width="1500" alt="ForgeCat Agent Profiles" src="./assets/forgecat_banner.png" /> </a>
<h3 align="center">Open catalog of cross-platform agent profiles</h3>
<div align="center"> <a href="https://forgecat.ai">Website</a> | <a href="https://forgecat.ai/docs">Documentation</a> | <a href="#quick-start">Quick Start</a> | <a href="#browse-profiles">Browse Profiles</a> | <a href="https://github.com/nota-america/forgecat-agent-profiles/issues/new?template=profile_request.yml">Request a Profile</a> </div> </div>
<br />
<div align="center">
</div>
<h3 align="center">Star the catalog to help developers discover reusable agent workflows.</h3>
ForgeCat Agent Profiles is the public catalog for installable agent profiles: skills, agents, rules, commands, hooks, and MCP settings packaged from useful open-source projects and made available across supported coding tools.
<h2 id="quick-start">Quick Start</h2>
Install the ForgeCat CLI:
npm install -g forgecatSearch the catalog:
forgecat search frontend
forgecat search mcp
forgecat search reviewInstall a profile:
forgecat install @forgecat/openai_skills_frontend-skill -p codex
forgecat install @forgecat/everyinc_compound-engineering-plugin -p claude-code
forgecat install @forgecat/obra_superpowers -p cursorUse forgecat install <profile> -p <platform> with one of the supported platform ids below.
Agent workflows are increasingly stored as project files: AGENTS.md, CLAUDE.md, .cursorrules, skills, subagents, commands, hooks, and MCP configuration. ForgeCat profiles make those workflows easier to discover, inspect, install, and port across tools.
| Need | ForgeCat profile repo provides |
|---|---|
| Find useful agent workflows | A curated catalog grouped by original source owner and repository |
| Reuse across coding tools | Platform-aware profile packages for Claude Code, Cursor, and Codex |
| Trust what is being installed | Source links, license metadata, profile READMEs, and conversion notes |
| Improve the catalog | Issue templates for profile requests, metadata fixes, platform support, and maintainer requests |
| Platform | ForgeCat id | What gets installed |
|---|---|---|
| Claude Code | claude-code | Claude-native skills, agents, commands, MCP, hooks, and project settings where supported |
| Cursor | cursor | Cursor rules, agents, MCP configuration, and compatible profile resources |
| Codex | codex | Codex skills, instructions, MCP configuration, hooks, and compatible profile resources |
Each profile declares its tested and partial platforms in for-forgecat/profile.yml. When a profile is marked tested, the matching native for-<platform> artifact should also exist in the profile directory.
ForgeCat profiles are curated to be source-faithful and reviewable, not just copied into a common folder shape.
| Signal | What it means |
|---|---|
| Source reviewed | The upstream repository was inspected directly before conversion. |
| Attribution kept | Profile metadata and docs link back to the original source repository. |
| License checked | The upstream license is reflected in for-forgecat/profile.yml when it is clear. |
| Install evidence | A platform is marked tested only after a real install/runtime check. |
| Native artifact included | Tested platforms should have matching for-claude/, for-cursor/, or for-codex/ directories. |
| Profile | Install | Good for |
|---|---|---|
| OpenAI frontend skill | forgecat install @forgecat/openai_skills_frontend-skill -p codex | Visually strong web apps, landing pages, prototypes, and UI polish |
| Compound Engineering | forgecat install @forgecat/everyinc_compound-engineering-plugin -p claude-code | Code review, research, planning, design review, and engineering workflows |
| Superpowers | forgecat install @forgecat/obra_superpowers -p cursor | Planning, TDD, debugging, review, subagents, and session-start workflow guidance |
| tldraw | forgecat install @forgecat/tldraw_tldraw -p cursor | Visual collaboration with agent workflows |
| Azure skills | forgecat search azure | Azure deployment, observability, compliance, RBAC, storage, and AI workflows |
Profiles are grouped by their original source owner and repository:
profiles/<source-owner>/<source-repository-or-collection>/<details> <summary>Show all collections</summary>
| Collection | Profiles | Description |
|---|---|---|
| addyosmani/agent-skills | 1 | Production-grade engineering skills pack by Addy Osmani |
| anthropics/claude-code | 1 | Claude Code profile conversion |
| anthropics/claude-plugins-official | 16 | Official Anthropic Claude plugin conversions for skills, agents, commands, and MCP |
| anthropics/knowledge-work-plugins | 15 | Knowledge-work plugins for legal, finance, HR, marketing, operations, sales, and more |
| anthropics/skills | 17 | Anthropic skill profiles for documents, spreadsheets, slides, web apps, themes, and more |
| awslabs/aidlc-workflows | 1 | AWS AI-DLC workflow rules for planning, construction, and governance |
| coreyhaines31/marketingskills | 1 | Marketing skills for conversion, copywriting, SEO, paid ads, analytics, and GTM work |
| cursor/plugins | 3 | Cursor plugins for team workflows, continual learning, and iterative agent loops |
| dietrichgebert/ponytail | 1 | Ponytail mode hooks and workflow support |
| egonex-ai/understand-anything | 1 | Codebase understanding skills for knowledge graphs, dashboards, architecture analysis, and onboarding |
| everyinc/compound-engineering-plugin | 1 | AI-powered development workflows for review, research, design, and delivery |
| fivetaku/fablize | 1 | Claude Code harness for grounded multi-step work and verification discipline |
| forgecat/forgecat-agent-profiles | 1 | ForgeCat-authored orchestration profile built from existing profile packages |
| garrytan/gstack | 1 | Garry's Stack engineering workflow skills for planning, QA, review, and deployment |
| google-gemini/gemini-skills | 3 | Gemini API, interactions API, and live API development skills |
| googleworkspace/cli | 1 | Google Workspace CLI skills for Drive, Gmail, Calendar, Docs, Sheets, Slides, Chat, and Apps Script |
| juliusbrussee/caveman | 1 | Caveman communication mode with compressed skills, commands, agents, and hooks |
| kepano/obsidian-skills | 1 | Obsidian vault, Markdown, Bases, Canvas, and CLI workflows |
| leonxlnx/taste-skill | 1 | Taste Skill package for premium frontend design and image-to-code workflows |
| mattpocock/skills | 1 | TypeScript skills by Matt Pocock |
| microsoft/azure-skills | 20 | Microsoft Azure skills for deployment, observability, compliance, storage, AI, and more |
| msitarzewski/agency-agents | 13 | Agency role agents for academic, design, engineering, marketing, sales, and operations teams |
| multica-ai/andrej-karpathy-skills | 1 | Andrej Karpathy coding guidelines for focused, verifiable agent work |
| obra/superpowers | 1 | Superpowers software-development methodology with hook-backed startup context |
| openai/skills | 45 | OpenAI skills for frontend, Figma, Playwright, deployment, documents, security, and more |
| orchestra-research/ai-research-skills | 1 | AI research skills for training, evaluation, inference, MLOps, RAG, multimodal systems, and papers |
| paramchoudhary/resumeskills | 1 | Resume and career-development skills for ATS optimization, resume writing, interview prep, and offer decisions |
| remotion-dev/skills | 1 | Remotion best-practices skill for video creation in React |
| tldraw/tldraw | 1 | Visual collaboration profile powered by the tldraw MCP server |
| upstash/context7 | 1 | Context7 MCP integration for up-to-date documentation lookup |
| vercel-labs/agent-skills | 7 | Vercel agent skills for deployment, React, UI audits, and workflow patterns |
| voltagent/awesome-codex-subagents | 10 | Codex subagents for development, infrastructure, security, and research |
| xixu-me/skills | 1 | Xi Xu skills for userscripts, GitHub Actions docs, secure hosting, browser workflows, and tooling |
| yeachan-heo/oh-my-claudecode_agents | 1 | Agents-only Claude Code orchestration profile from oh-my-claudecode |
</details>
Total: 173 profiles across 34 collections.
A complete profile directory usually contains:
profiles/<source-owner>/<source-repository>/
├── README.md
├── for-forgecat/
│ ├── profile.yml
│ └── ...
├── for-claude/
├── for-cursor/
└── for-codex/for-forgecat/ is the canonical ForgeCat package source. Native for-claude, for-cursor, and for-codex directories are added when the profile has been tested and materialized for those platforms.
Contributions are welcome when they improve installability, source fidelity, documentation, or platform support.
Read CONTRIBUTING.md before opening a PR.
Good contributions include:
Before opening a PR:
profiles/<source-owner>/<source-repository-or-collection>/.forgecat validate from the profile's for-forgecat/ directory.for-<platform> artifact and note the install/runtime evidence in the PR.If you maintain an upstream repository that has been converted here, use the maintainer request issue template for corrections, takedown requests, or ownership changes. We can update attribution, fix platform notes, remove a profile, or help you submit the profile directly.
Open an issue with:
claude-code, cursor, codex, or all supported platformsProfiles with clear agent surfaces, permissive licensing, active usage, and strong developer value are the best candidates.
npm install -g forgecatThis repository is distributed under the Apache License 2.0. See LICENSE.
Individual profiles may include content from their original source repositories and can carry their own license terms. Check each profile's for-forgecat/profile.yml, README, and included license files before redistributing or modifying profile content.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.