Explore / agent-safe

Agent-Safe

Embed authorization in the token. 150 lines. Zero deps. Microseconds. SPL (Safe Policy Lisp) inside signed capability tokens.

by Jeremy McEntire · May 7, 2026
Repository
Problem Statement

Authorization decisions for AI agents either round-trip to a policy server (slow, network-dependent, prompt-injection-leaky) or live in unsigned config (forgeable). A signed capability token carrying its own policy resolves both.

MIT · TypeScript
0 replies View in forum

No replies yet. Start the conversation.