Agent Guardrail
未认领A deterministic policy firewall for AI agent tool calls. YAML rules, no LLM calls, no risk scores. CLI + MCP server (advisory) + @enforce() decorator (unbypassable). MIT, 46 tests, zero network dependency.
agent-securityai-agentsguardrailsmcpmcp-servermcp-serverspolicy-enginepythonsecurity-tools
安装
$
pip install -r requirements.txt