guardclaw

7-layer runtime security for ai agents. local-first, deterministic security layer that blocks risky actions before they run. no llm in the security path.

Visit Website

More Security Tools

agentkeys

Security

give ai agents secure access to any service — through proxy tokens, not real credentials.

clawsecurity

Security

security scanning and remediation platform for ai agents and code

clawshell

Security

agent execution must be treated as adversarial compute, so we built clawshell, linux permissions for ai agent secrets. apache 2.0 licensed, written in rust