The authorization kernel for AI agents. Block, allow, or escalate agent tool calls with YAML rules — deterministic-first, LLM fallback.
python cli open-source security typescript policy openai ai-safety ai-agents guardrails llm langchain anthropic agent-authorization
-
Updated
Apr 4, 2026 - TypeScript