ClawSec

Security audit platform for MCP servers and AI agent skills. 5-tier analysis pipeline — static analysis, pattern matching, LLM semantic review, Firecracker sandbox execution, and LLM audit — detects malicious patterns, data exfiltration, and prompt injection. Database of 30,000+ audited skills with Trust Scores. Companion tool [ClawSearch](https://clawsearch.cc) provides safe skill discovery with security ratings.

Browse this MCP server in the TensorBlock MCP Index.