Deterministic policy firewall for AI agent tool calls - YAML rules, not a fuzzy risk score.
Free
2 servers across ai & agents. The largest is Agent Guardrail.
AI & Agents
2
servers published
1 category
4k
downloads
counted by the registries
0
followers
told when a release lands
2 of 2
security-scanned
servers with a scan grade
2 MCP servers, most downloaded first. 2 of 2 are free to start.
Explainable ALLOW/WARN/BLOCK risk decisions for AI agent actions before execution, signed as OAA.
Free
Authentication
0 of 2 use OAuth
The rest either hold an API key issued by the publisher or ask for nothing.
Support
Community publisher
Listed without a verification tick. Read the listing before installing.