tech-leads-club/agent-skills

The secure, validated skill registry for professional AI coding agents. Extend Antigravity, Claude Code, Cursor, Copilot and more with absolute confidence.

What it solves

Agent Skills provides a secure, validated registry of capabilities for AI coding agents. It addresses the security risks associated with open marketplaces, where a significant portion of skills can contain critical vulnerabilities, by providing a hardened library of verified and tested instructions.

How it works

The project consists of a CLI tool and an MCP server. The CLI allows users to browse, install, and update skills—which are packaged instructions, templates, and references—directly into the configuration of supported AI agents (such as Cursor, Claude Code, and GitHub Copilot). The MCP server enables agents to search and fetch these skills on-demand via progressive disclosure.

Who it’s for

Developers using AI coding agents who want to extend their assistants' workflows, patterns, and specialized knowledge with safe, curated capabilities.

Highlights

  • Hardened Security: Uses static analysis, Snyk Agent Scan, and human curation to ensure skills are safe.
  • Broad Agent Support: Compatible with a wide range of popular and enterprise AI agents including Claude Code, Cursor, Windsurf, and Amazon Q.
  • Flexible Installation: Supports global or local project installation via copy or symlink.
  • MCP Integration: Includes a dedicated MCP server for agents to dynamically discover and load skills.

Related

  • Project
  • Project
  • Project
  • Project
  • Project