Skill Registry
DirectoryA searchable directory of the best Claude skills from GitHub — with GitHub stars, a health score and an AI security check. Currently 6 skills.
Open →
Anthropic
Claude
Claude is Anthropic's AI assistant — as a chat, in Claude Code on the command line, and as an agent that operates real tools through Skills and the Model Context Protocol (MCP). This is where we gather what Claude can do today — and the Beckmann building blocks for it.
- 6 skills in the directory
- 589k GitHub stars total
- MCP server to plug in
- Command Guard for Claude Code
Building blocks for Claude
Three ways to make Claude more useful day to day — from the skill directory to the safety check.
Ready-made MCP tools to plug in instantly: knowledge-base search, news, command guard. One URL for everything: mcp.beckmann.ai.
Open →
Command Guard
SecurityChecks your AI agent's shell commands before they run — blocks destructive ones like rm -rf or git push --force.
Open →
What are Claude Skills?
Skills are folders of instructions, scripts and resources (a SKILL.md plus extras) that Claude loads on
demand to complete a specific task reliably — understand a video, review code, build documents. Most of them come from
open GitHub repositories. The Beckmann Skill Registry pulls that data, computes a health
score and checks for typical security risks with GPT-4o-mini.