CompletionKit for Claude.ai
com.completionkit/evals
Prompt evals over MCP: run a prompt on your dataset, score each output 1-5 with an LLM judge.
client:Claude.ai
transport:streamable-http
tools:54
Install CompletionKit in Claude.ai
Settings → Connectors → Add custom connector
https://completionkit.com/mcpClaude on the web takes a remote URL and completes OAuth in the browser. For a packaged server use Claude Desktop or Claude Code instead. Claude.ai docs
What Claude.ai can do once it is connected
- prompts_list
- prompts_get
- prompts_create
- prompts_update
- prompts_delete
- prompts_publish
- prompts_suggest_improvement
- runs_list
- runs_get
- runs_create
- runs_update
- runs_delete
- runs_generate
- runs_regrade
- runs_rerun
- runs_retry_failures
- responses_list
- responses_get
- datasets_list
- datasets_get
- datasets_create
- datasets_update
- datasets_delete
- datasets_create_from_url