crit for Grok
io.github.crit-labs/crit
Build, evaluate and serve a bespoke LLM inference engine for your own GPU with crit's commands.
client:Grok
transport:stdio
runtime:pypi
Install crit in Grok
.mcp.json (in your project root)
{
"mcpServers": {
"crit": {
"type": "stdio",
"command": "uvx",
"args": [
"crit"
]
}
}
}Grok Build reads Claude Code's .mcp.json and Claude Desktop's config unchanged, so a server you have already set up there needs nothing new. The Grok web app and X integration do not take custom servers. Grok docs