What it does
The publisher describes Agent Action Gate as a gate that denies destructive agent tools when they run unattended, with a "Prove" function also named in its description. The source does not explain how either function works.
The resource is listed under development and security, so it may suit developers who want to limit which agent tool calls can run without supervision. The available information does not describe specific integrations or supported agent platforms.
It is an MCP server, and the source gives no setup steps, configuration details or runtime requirements beyond its repository link.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
Agent Action Gate
Gate/Prove: deny unattended destructive agent tools. Instant Audit $499 on a2zsoc.com.
v0.2.0As shown on registry.modelcontextprotocol.io when we last read it