Heads up This site is currently under heavy development.
← all datasets
◆ Evals & benchmarks · MCP servers

Salesforce MCP-Universe

Apache-2.0 Runs code

A benchmark of agent tasks against real-world MCP servers across several domains, plus the framework to run and trace them.

source ↗ ★ 600 last push 2026-06-23

Summary

📊 MCPMark Evaluation - MCP-Universe now supports evaluating the MCPMark tasks 🚀 MCP+ - Agentic wrapper on MCP clients which reduce token costs by up to 75% 🔬 Deep Research Agent - Scale the Width of Deep Research Agents with parallel tool calling, improving performance and efficiency
from the project README
Formats
yaml
License
Apache-2.0
Added
2026-09-10

Get it

From the project’s own instructions where it documents any; otherwise a plain clone.

  1. shell 
    $ git clone https://github.com/SalesforceAIResearch/MCP-Universe.git

    Read the scripts it bundles before installing.

  2. shell 
    $ cd MCP-Universe && pip install -r requirements.txt

    Benchmarks call real MCP servers (maps, GitHub, finance, browser), so each needs its own keys in .env.

Contents

1,356 files 4.3 MB repository

  • .py 602
  • .json 492
  • .md 155
  • .yaml 31
  • .txt 23
  • .sql 13
  • .j2 8
  • (no ext) 8
my-toolchain — 0 tools
paste an install list to detect your tools

A brew list, a Brewfile, requirements.txt, a Dockerfile — or just the product names, free-form. Nothing leaves your browser.

    browse all tools →

    Send us feedback