"author:batrapulkit" matching MCP servers:

Enables AI agents and MCP clients to route every tool call through configurable rules that allow, block, or hold risky actions for human approval, while recording decisions in a tamper-evident audit trail.