agent guardrail for Claude Code

io.github.rudimentall1/agent-guardrail

Deterministic policy firewall for AI agent tool calls - YAML rules, not a fuzzy risk score.

client:Claude Code transport:stdio runtime:pypi

Install agent guardrail in Claude Code

run in your project directory

claude mcp add agent-guardrail -- uvx guardrail-mcp

Adds it for this project only. Append --scope user to make it available everywhere. Claude Code docs