MCP.so
Sign In
U

UniversalBench: run code, web, databases, and any LLM from your AI, safely

@nikhilgogulwar

Repository

About UniversalBench: run code, web, databases, and any LLM from your AI, safely

An MCP execution platform that gives any AI real infrastructure to act. Through three tools it can run code and shell commands, install packages, run tasks in parallel, search the web, make HTTP calls, read write and query databases, call any LLM, commit to GitHub, take screensho

#mcp#code-execution#ai-agents#llm#database#github#web-search#developer-tools

Basic information

Category

Developer Tools

Transports

stdio

Publisher

nikhilgogulwar

Submitted by

nikhilgogulwar

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "universalbench": {
      "url": "https://universalbench-mcp.penantiaglobal.workers.dev/u/{api_key}"
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is UniversalBench?

UniversalBench is an MCP server that gives AI the infrastructure to act by connecting one URL to Claude, ChatGPT, Gemini, or any MCP-compatible AI. It lets AI run code, search the web, read and write databases, call any LLM, and commit to GitHub, filtering results before they reach the model to cut up to 96.5% of tokens versus doing the work in chat.

How to use UniversalBench?

Connect via the URL https://universalbench-mcp.penantiaglobal.workers.dev/u/{api_key} with any MCP-compatible AI. Your AI then gets three tools — ub_read, ub_write, and ub_ai — to perform actions in an isolated sandbox. A free tier offers 1,000 calls per month with no card required, then pay as you go.

Key features of UniversalBench

  • Run code and shell commands in an isolated sandbox
  • Search the web and make safe HTTP calls
  • Read, write, and query databases with your own credentials
  • Call any major LLM with cost caps on by default
  • Read from and commit to GitHub with every push validated first
  • Store credentials in an encrypted vault that unlocks tools automatically

Use cases of UniversalBench

  • AI running and testing code in a secure sandbox
  • AI searching the web and fetching external data
  • AI reading and writing to your own databases
  • AI calling any major LLM with automatic cost caps
  • AI committing validated code directly to GitHub

FAQ from UniversalBench

What guarantees does UniversalBench provide?

Three live guarantees: AI never ships broken code, never burns your wallet with uncapped costs, and cannot reach your internal network.

How does UniversalBench reduce token usage?

By filtering results before they reach the model, it cuts up to 96.5% of tokens compared to doing the same work inside chat.

What tools does UniversalBench expose to AI?

It exposes three tools: ub_read, ub_write, and ub_ai.

Does UniversalBench have a free tier?

Yes, it offers 1,000 calls per month with no credit card required, then pay as you go.

How do I connect to UniversalBench?

Connect any MCP-compatible AI via the URL `https://universalbench-mcp.penantiaglobal.

Comments

More Developer Tools MCP servers

test

prysmaticlabs

Go implementation of Ethereum proof of stake

3.8K

JetBrains MCP Proxy Server

JetBrains

A model context protocol server to work with JetBrains IDEs: IntelliJ, PyCharm, WebStorm, etc. Also, works with Android Studio

962

Test

x1xhlol

FULL Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI, VSCode Agent, Warp.dev, Windsurf, Xcode, Z.ai Code, Dia & v0. (And other Open Sou

141.7K

Hello World MCP Server (Reference Extension)

anthropics

Desktop Extensions: One-click local MCP server installation in desktop apps

2.0K

Stakpak Agent CLI

stakpak

Ship your code, on autopilot. An open source agent that lives on your machines 24/7 and keeps your apps running. 🦀

1.6K