MCP Servers

@pathintegral-institute

Repository Homepage

Basic information

About MCP Servers

Open Source MCP Servers for Scientific Research

#mcp-servers #open-source#scientific-research

MCP Servers config

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

{
  "mcpServers": {
    "mcp-servers": {
      "command": "uvx",
      "args": [
        "mcp-science",
        "web-fetch"
      ]
    }
  }
}

MCP Servers tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

Fetch the live tool list by running this server in a temporary sandbox using the button above.

Overview

What is MCP Servers?

MCP Servers is a monorepo collection of open-source MCP (Model Context Protocol) servers designed for scientific research applications. They enable AI models to interact with scientific data, tools, and resources through a standardized protocol.

How to use MCP Servers?

Install uv and an MCP‑enabled client (e.g. Claude Desktop, VSCode, Goose, 5ire). Launch any server with uvx mcp-science <server-name> (e.g. uvx mcp-science web-fetch). Optionally use the mcpm tool to automate client configuration.

Key features of MCP Servers

Open-source collection built for scientific research
Standardized MCP protocol for AI–data integration
Single‑command launch via uvx mcp-science
Covers materials science, web fetch, Python execution, SSH, DFT, and more
Works with multiple MCP‑enabled clients (Claude Desktop, VSCode, Goose, 5ire)
Packaged as a Python monorepo on PyPI (mcp-science)

Use cases of MCP Servers

Search and visualise materials‑science data from the Materials Project
Fetch and summarise web content (HTML, PDF, plain text)
Execute Python code in a sandboxed environment for safe analysis
Run pre‑validated commands on remote machines over SSH
Perform density‑functional‑theory (DFT) calculations via GPAW
Interact programmatically with a running Jupyter kernel

Frequently asked questions

What is MCP?

MCP is an open protocol that standardises how applications provide context to LLMs, similar to a USB‑C port for AI. It allows models to integrate with various data sources and tools in a consistent way.

What are the prerequisites for using MCP Servers?

You need `uv` (a fast Python package manager) and an MCP‑enabled client such as Claude Desktop, VSCode, Goose, or 5ire.

How do I run a server?

Use the command `uvx mcp-science <server-name>` (e.g. `uvx mcp-science web-fetch`). This downloads the `mcp-science` package from PyPI and launches the server.

Can I build my own MCP server?

Yes. The repository includes a step‑by‑step guide under `docs/how-to-build-your-own-mcp-server-step-by-step.md` and an example server in `servers/example-server/`.

How can I contribute to MCP Servers?

Fork the repository, create a feature branch, make changes with clear commit messages, update documentation, add tests, and open a pull request. Follow the naming conventions described in the contributing guide.

Comments

More Other MCP servers

Awesome Mlops

visenger

A curated list of references for MLOps

14.0K

Screenpipe

screenpipe

YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context

22.0K

EverArt

modelcontextprotocol

Model Context Protocol Servers

88.1K

Hindsight

vectorize-io

Hindsight: Agent Memory That Learns

50.4K

Volley.sh

Volley

A good sequence diagram can solve almost any communication problem in engineering. Volley lets your agent write them as plain text, with the real request and response payloads on every arrow. It checks its own syntax wi