AgentGuard
by Carlos Montiel
Free
Local security scanner for AI agent skills and MCP servers: tool poisoning, rug pulls, known CVEs, the lethal trifecta, and an experimental AI semantic check (--deep).
1
0
I build security tools for AI agents. Creator of agentguard, an open-source scanner that catches tool poisoning, rug pulls, known CVEs and data-exfiltration risks in skills and MCP servers. Founder of The Only Edge (Vanguard Growth LLC), Houston. Open to security audits for teams running agent skills.