Nonobench

io.github.mauricekleine/nonobench

An open-source benchmark of how well LLMs solve nonogram puzzles, from 5x5 to 20x20.

Description as published in the official MCP registry.

Answers the MCP handshake and lists its tools. Checked 2026-10-04.

What we measured

Endpointhttps://www.nonobench.com/mcpregistry
HandshakeAnswers the MCP handshake and lists its tools (599 ms)our check, 2026-10-04
Protocol version2026-07-28our check, 2026-10-04
Tools (11)get_leaderboard, list_providers, list_families, compare_models, get_model_results, list_puzzles, get_puzzle, check_solution, get_puzzle_results, get_model_puzzles, list_runstools/list, 2026-10-04
GitHub stars6github.com/mauricekleine/nonobench
Last push2026-10-01GitHub API
LicenceMITGitHub API
Registry version1.1.0 · active · updated 2026-09-29registry

Signed reports

No signed reports yet. A report carries a signed decision record from the reporter's gate, so it shows a real call went through, not just an opinion. How to file one.

Connect

claude mcp add --transport http nonobench https://www.nonobench.com/mcp

Commands are built from the registry entry. Check the publisher's documentation before giving any server access to your data.

Badge for your README

Shows the result of our latest check and links back to this page.

[![agora](https://openforallofus.com/badge/io.github.mauricekleine/nonobench.svg)](https://openforallofus.com/tools/io.github.mauricekleine/nonobench)

Agents can read this page as data: MCP endpoint https://openforallofus.com/api/mcp, tool get_tool with name io.github.mauricekleine/nonobench.