garagon
No LLM, no cloud, no API keys. One Go binary, scans in milliseconds.
Four analysis layers: 1. Pattern matching (regex + base64/hex decoding) 2. NLP markdown analysis (AST walker, keyword classification) 3. Taint tracking (source-to-sink flow analysis) 4. Rug-pull detection (hash tracking across scans)
Every rule includes remediation guidance - when it finds a problem, it tells you how to fix it.
We also run Aguara Watch (watch.aguarascan.com), scanning 45,000+ skills across 7 public registries daily. The data shapes the rules.
Install: brew install garagon/tap/aguara Docker: docker run ghcr.io/garagon/aguara scan /scan GitHub Action: uses: garagon/aguara@v1
From Skill.md to Shell: A Security Audit Guide for AI Agent Skills
Show HN: Aguara – Security scanner for AI agent skills and MCP servers
https://github.com/garagon/aguara
Hey HN, I built Aguara because I kept seeing the same problem: AI agents and MCP servers run code on your behalf, and nobody is checking what that code actually does before it runs. A single malicious skill file can exfiltrate your SSH keys, inject prompts to override safety instructions, or curl-pipe-bash a backdoor. I wanted something like Semgrep but specifically for the AI...