{"id":"mcp-bench","name":"MCP-Bench","maker":"Accenture","url":"https://github.com/Accenture/mcp-bench","repo":"Accenture/mcp-bench","summary":"A benchmark of tool-using agent tasks that run against a collection of MCP servers.","details":"Tasks need agents to plan and use tools from many MCP servers, covering fuzzy instructions and cross-tool coordination, and results are scored with rubrics and a judge model. The code is published on GitHub.","open_source":true,"hosted":false,"speaks":["mcp"],"verified":"2026-11-15","category":"benchmarks","category_name":"Agent benchmarks","page":"https://modelmoot.com/tools/mcp-bench","reviewed":"2026-11-15","checked":{"at":"2026-11-15T15:07:43.409Z","address_answers":true},"note":"Descriptions are the site's own. Facts under \"checked\" were read automatically from the tool's address and public repository."}