openagentidentityprotocol/agentidentityprotocol

View on GitHub ↗

Agent Identity Protocol - Zero-trust security layer for AI agents. Policy enforcement proxy for MCP with Human-in-the-Loop approval, DLP scanning, and audit logging.

36 ★4 forksUpdated 6mo ago

What you need to know

Standard proposal for Agent Identity Protocol, a zero-trust IAM layer for agent-to-agent authentication and authorization, rather than a runnable MCP server.

Reviewed 2026-08-11

Topics

agent-identity-protocolai-agentsai-safetycursor-idedlpgolanghuman-in-the-loopllmmcpmodel-context-protocolpolicy-enforcementsecurityzero-trust
Stars
36★
Forks
4
Language
License
Apache-2.0
Created
2026-01-20
Last push
2026-03-05