What it does
According to the publisher, evalgate provides statistical checks that an agent can run on an AI evaluation result before relying on it. The publisher lists examples including whether a number-one ranking is real and whether a judge model is biased, and mentions further checks without naming them.
The description suggests use when an agent or developer needs to verify an AI eval score before acting on it. The source does not describe setup or usage steps.
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/evalgate
Statistical checks an agent runs before trusting an AI eval number (is #1 real, judge bias, more).
v0.4.1As shown on registry.modelcontextprotocol.io when we last read it