aiskillstore/marketplace
Security-audited skills for Claude, Codex & Claude Code. One-click install, quality verified.
What it solves
This project provides a centralized, open-source repository for approved AI agent skills. It solves the problem of discovering, distributing, and verifying the security of specialized capabilities (skills) that can be added to AI tools like Claude Code and Codex.
How it works
The repository acts as a backend storage for the Skill Store platform. Skills are submitted via a web interface, undergo an automated security audit to check for dangerous code patterns, network calls, or credential leaks, and are then approved by maintainers. Once approved, they are stored here and can be installed via a dedicated CLI tool (npx skillstore add) which places the skill files into the correct directories for the target AI tool.
Who it’s for
Developers and users of AI agents (specifically Claude Code and Codex) who want to extend their agent's capabilities with pre-verified, community-contributed skills.
Highlights
- Automated Security Audits: Every skill is scanned for risks like
evalcalls, unauthorized file access, and obfuscated code before publication. - Automated Pipeline: The repository is managed by an automated pipeline that handles submissions, audits, and synchronization.
- CLI Installation: A simple command-line interface allows for quick installation of skills directly into the agent's environment.
- Standardized Specification: Based on the Agent Skills specification to ensure compatibility across different tools.
Related
- Project
- Project
- Project
- Project
- Project