Score text, files and pages against a rubric with decision models; compare models on your cases.
Maintain this server? Claim it
This server only runs locally, from a package. Install it with the instructions on this page to try it.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add verdict \ -- npx -y @verdict-ai/[email protected]
Environment
TYPESAFE_API_KEY (optional, not in the config above): Key for Jev, the default decision model, when judging runs in this process.CLOUDFLARE_ACCOUNT_ID (optional, not in the config above): Cloudflare account for Clef and Clef-flash on Workers AI.CLOUDFLARE_API_TOKEN (optional, not in the config above): Cloudflare token (Workers AI: Read) for Clef and Clef-flash.OPENAI_API_KEY (optional, not in the config above): Key for OpenAI's Decisions API.VERDICT_DATA_DIR (optional, not in the config above): Folder to keep rubrics and evaluations in between runs. Without it they are kept in memory.VERDICT_API_URL (optional, not in the config above): Address of a Verdict API to use instead of judging in this process.VERDICT_API_KEY (optional, not in the config above): Key for that Verdict API.We don't run packages, so we can't list this server's tools.