wellworn
catalog entry

CompletionKit

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

website completionkit.comsource github.com
signals
stars
n/a
commits 3mo
n/a
last release
1.0.02026-07-18
status
active

Stars count attention, not fitness. The release date and the three-month commit count say more about whether anyone is still maintaining this.

No Wellworn verdict has been tested against CompletionKit yet, so nothing on this page is a recommendation.

licence
not recorded
categories
mcp

All categories