We are building Caracal, an open-source trust and security layer for autonomous AI agents.
As agents gain access to APIs, MCPs, databases, internal tools, credentials, and sensitive workflows, organizations need a way to control what agents can do, who authorized those actions, how permissions are delegated, and how every action is audited. That’s the problem Caracal is designed to solve.
Over the last several months, we’ve focused on security reviews, testing, and improving the platform, and are now opening it up to the broader community. Along the way, we’ve been fortunate to receive support from GitHub Secure Fund, Microsoft for Startups**, Founders Inc. Canopy, Vercel Open Source Program, and LFX Mentorship** under the Linux Foundation Decentralized Trust umbrella.
Repo: GitHub - Garudex-Labs/caracal: 🐾 Caracal is a security-first AI authority, delegation, and identity system for autonomous agents. · GitHub
Community: Caracal