luongnv89/asm

The universal skill manager for AI coding agents.

What it solves

agent-skill-manager (asm) solves the problem of fragmented skill management for AI agents. Instead of manually copying skill files into different hidden directories for tools like Claude Code, Cursor, and Windsurf, it provides a centralized CLI to install, search, audit, and organize these skills across multiple providers from a single interface.

How it works

It acts as a management layer between skill repositories (GitHub or the ASM Registry) and the specific configuration folders of various AI agent tools. The tool validates the SKILL.md format (YAML frontmatter and markdown body), performs security scans for malicious patterns, and then deploys the skills to the correct provider paths. It is designed to be scriptable, offering JSON output and non-interactive flags for integration into AI agent workflows or CI pipelines.

Who it’s for

  • AI Agent Users: People using multiple AI coding assistants (e.g., Claude Code, Cursor, Copilot) who want to synchronize and manage their custom instructions and skills.
  • Skill Authors: Developers creating reusable agent skills who need tools for scaffolding, local testing via symlinking, and publishing to a registry.
  • Automation Engineers: Those building agentic workflows that need to programmatically manage the capabilities of their AI tools.

Highlights

  • Cross-Provider Support: Manages skills for 19 different agents, including Claude Code, Cursor, Windsurf, and GitHub Copilot.
  • Agent-Ready CLI: Supports --json and --yes flags for machine-readable output and non-interactive execution.
  • Security Auditing: Scans skills for shell execution, network access, and credential exposure before installation.
  • Centralized Registry: Access to a catalog of over 4,300 community-published skills.
  • Developer Pipeline: Includes commands for scaffolding (init), live development (link), and quality evaluation (eval).