← Top 100 / Full directory (14,309)

CompletionKit

AI & Memory Remote Official

Prompt evals over MCP: run a prompt on your dataset, score each output 1-5 with an LLM judge.

GitHub ↗ Docs ↗

How to connect

1
Add the remote MCP URL
Paste this endpoint into Claude, ChatGPT, or Cursor MCP settings.
https://completionkit.com/mcp
2
Verify the tools loaded
Confirm tools/list returns the expected tools before relying on them in production.

Tools

Tool names are not listed in our registry for this server. Connect it in your MCP client to see the live tool list.

Related servers