Evalgate runs on your own machine — the client starts it, so there is no endpoint to ping. 1 installs a week from pypi.
Statistical checks an agent runs before trusting an AI eval number (is #1 real, judge bias, more).
The linked repository no longer exists on GitHub — it was deleted or made private.
This one has been quiet for a while
Quiet is not dead — but it is worth knowing when it wakes up, or when someone else takes it over. We watch the repository and tell you either way.
Connect this server
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add evalgate -- uvx eval-integrity{
"mcpServers": {
"evalgate": {
"command": "uvx",
"args": [
"eval-integrity"
]
}
}
}[mcp_servers.evalgate]
command = "uvx"
args = ["eval-integrity"]{
"mcpServers": {
"evalgate": {
"command": "uvx",
"args": [
"eval-integrity"
]
}
}
}{
"servers": {
"evalgate": {
"type": "stdio",
"command": "uvx",
"args": [
"eval-integrity"
]
}
}
}Alternatives to Evalgate
same job, measured the same wayCheck an agent's INAM reputation before trusting it, and sign a receipt when work is done.
Read-only PostgreSQL for AI agents: every statement is checked before it runs. No account needed.
An AI agent's trading mandate it can't break: checked before every order, reasoning sealed in IRL.
MCP proxy that checks every tools/call against Agent Trust policy, with evidence, before it runs.
Check GitHub repos and shell commands for malware-delivery tricks before an AI agent runs them
See which AI models fall for traps in real jobs, and check MCP servers or skills before install.
Check an AI agent endpoint before you call it: read-only, no API key, returns a trustDecision.
Tanilo checks the premise before an AI agent acts.
Evalgate — questions
Answers built from our own checks of this server.