Show HN: Guarden – Authorization for AI agent actions powered by OPA

Category: security

Tags: authorization, ai-agents, security, opa, policy-as-code

Score: 7.0/10 (Innovation: 7, Technical: 8, Documentation: 7, Utility: 6)

Guarden is a proof-of-concept authorization layer for AI agents that enforces policies at the action boundary rather than the SDK, using OPA (Open Policy Agent) to normalize actions from multiple paths including code execution. Its innovative approach of separating authorization from the transport layer and generating least-privilege Rego policies from tool schemas addresses a critical gap in AI agent security.

Target audience: backend devs, security engineers, AI developers

Repository: https://github.com/las7/Guarden · Python · Apache-2.0 · 1 stars

View on Hacker News