What it does
The server, published by ipezygj as io.github.ipezygj/agent-guard, provides checks an AI agent can run before it takes an action. According to the publisher, these cover three areas: whether a package may contain malware, whether a command would be destructive, and whether a secret could leak.
It suits developers who give an agent the ability to install packages, run shell commands or handle credentials, and who want a screening step before those actions go ahead. The publisher's description does not detail how each check is performed or what it reports back.
The source does not describe setup steps, runtime requirements or the exact tools the server exposes. The project is hosted on GitHub.
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
io.github.ipezygj/agent-guard
Safety checks an agent runs before it acts: package malware, destructive command, secret leak.
v0.2.1As shown on registry.modelcontextprotocol.io when we last read it