MCP.science: Open Source MCP Servers for Scientific ResearchCAUTION
Open Source MCP Servers for Scientific Research
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://opensource.org/licenses/MIT)
Join us in accelerating scientific discovery with AI and open-source tools!
Quick Start
Running any server in this repository is as simple as a single command. For example, to start the web-fetch server:
uvx mcp-science web-fetch
This command handles everything from installation to execution. For more details on configuration and finding other servers, see the "How to configure MCP servers for AI client apps" section below.
Table of Contents
- About
- What is MCP?
- Available servers in this repo
- How to integrate MCP servers into LLM
- How to build your own MCP server
- Contributing
- License
- Acknowledgments
- Citation
About
This repository contains a collection of open source MCP servers specifically designed for scientific research applications. These servers enable Al models (like Claude) to interact with scientific data, tools, and resources through a standardized protocol.
What is MCP?
MCP is an open protocol that standardizes how applications provide context to LLMs. Think of MCP like a USB-C port for AI applications. Just as USB-C provides a standardized way to connect your devices to various peripherals and accessories, MCP provides a standardized way to connect AI models to different data sources and tools. MCP helps you build agents and complex workflows on top of LLMs. LLMs frequently need to integrate with data and tools, and MCP pro
7168561a9c98OBSERVED · 2026-10-07Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add web-fetch --env JINA_API_KEY=${JINA_API_KEY} --env JUPYTER_SERVER_TOKEN=${JUPYTER_SERVER_TOKEN} --env MP_API_KEY=${MP_API_KEY} --env NEMAD_API_KEY=${NEMAD_API_KEY} -- uvx web-fetch{
"mcpServers": {
"web-fetch": {
"command": "uvx",
"args": [
"web-fetch"
],
"env": {
"JINA_API_KEY": "${JINA_API_KEY}",
"JUPYTER_SERVER_TOKEN": "${JUPYTER_SERVER_TOKEN}",
"MP_API_KEY": "${MP_API_KEY}",
"NEMAD_API_KEY": "${NEMAD_API_KEY}"
}
}
}
}Exposed tools (42)
24 read · 14 write · 4 destructive. Blast radius: 4 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
add | write | Add two numbers. |
analyze_eigenstate | read | Analyze a specific eigenstate of a quantum system. |
analyze_ground_state | read | Analyze the ground state properties of a quantum system. |
build_supercell | read | |
check_calculation_result | read | |
compute_energy_spectrum | read | Compute the energy spectrum of a quantum system. |
create_new_project | write | |
create_quantum_system | write | Create a new quantum system for analysis. |
create_structure_from_cif | write | |
create_structure_from_poscar | write | |
create_table | write | Creates a new table in the database. |
delete_documents | destructive | Deletes documents from the specified table. |
delete_quantum_system | destructive | Delete a quantum system and its associated data. |
display_numerical_result | read | Displays a saved plot for a given quantum system analysis. |
drop_table | destructive | Drops the specified table from the database. |
execute_mathematica | write | |
generate_plot | read | Generate plots for quantum system analysis. |
get_image_of_flower | read | Get an image of flower |
get_system_details | read | Get detailed information about a specific quantum system. |
insert_document | write | Inserts a document into the specified table. |
list_quantum_systems | read | List all available quantum systems. |
moire_homobilayer | read | |
nemad_formula_search | read | Search for materials by exact chemical formula in NEMAD database. |
nemad_list_cache | read | List all available cached search results. |
nemad_read_results | read | Read specific range of cached search results. |
nemad_search | read | Search for materials by elements in NEMAD database. |
parameter_sweep | read | Perform a parameter sweep for a quantum model. |
plot_structure | read | |
plot_xy | write | Generate a generic 2D plot and save it to a file. |
purge_table | destructive | Removes all documents from the specified table. |
query_documents | read | Queries documents from the specified table. |
read-image-url | read | Read images from URLs, convert to base64, and return the image content. |
search_materials_by_formula | read | |
select_material_by_id | read | |
set_hamiltonian | write | Set the Hamiltonian for a quantum system with optional parameter ranges for sweeps. |
set_hilbert_space | write | Set the Hilbert space for a quantum system. |
set_lattice | write | Set the lattice geometry for a quantum system. |
start_calculation | write | |
txyz_search_smart | read | return await _search( |
txyz_search_web | read | return await _search( |
update_documents | write | Updates documents in the specified table. |
verify_derivation | read |
Trust audit
CAUTIONgrade B · trust 86/100 Install with care. The audit found things worth knowing before you trust its output.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | WARN |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (5 observation(s))
- Network
- declared (5 observation(s))
- Shell
- declared (1 observation(s))
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (18)
raw_module = __import__(expression.module, fromlist=[
image_base64 = "iVBORw0KGgoAAAANSUhEUgAAAB4AAAAeCAYAAAA7MK6iAAAAIGNIUk0AAHomAACAhAAA+gAAAIDoAAB1MAAA6mAAADqYAAAXcJy6UTwAAAAGYktHRAD/AP8A/6C9p5MAAAAJcEhZcwABGUAAARlAAYDjddQAAAAHdElNRQfpBBUNAgfLUoX1AAAA
delete_documents, delete_quantum_system, drop_table, purge_table
.dxtignore
.all-contributorsrc
.dxtignore
.dxtignore
.all-contributorsrc
DEV_SERVERS_PATH = (Path(__file__).parent / '../../servers').resolve()
image_base64 = "iVBORw0KGgoAAAANSUhEUgAAAB4AAAAeCAYAAAA7MK6iAAAAIGNIUk0AAHomAACAhAAA+gAAAIDoAAB1MAAA6mAAADqYAAAXcJy6UTwAAAAGYktHRAD/AP8A/6C9p5MAAAAJcEhZcwABGUAAARlAAYDjddQAAAAHdElNRQfpBBUNAgfLUoX1AAAA
image_base64 = "iVBORw0KGgoAAAANSUhEUgAAAB4AAAAeCAYAAAA7MK6iAAAAIGNIUk0AAHomAACAhAAA+gAAAIDoAAB1MAAA6mAAADqYAAAXcJy6UTwAAAAGYktHRAD/AP8A/6C9p5MAAAAJcEhZcwABGUAAARlAAYDjddQAAAAHdElNRQfpBBUNAgfLUoX1AAAA
### 5ire (open source, need AI API key, macOS / Linux / Windows)
### Goose Desktop (open source, need AI API key, macOS / Linux)
curl -sSf https://astral.sh/uv/install.sh | bash
curl -LsSf https://astral.sh/uv/install.sh | sh
curl -LsSf https://astral.sh/uv/install.sh | sh
curl -LsSf https://astral.sh/uv/install.sh | sh
curl -LsSf https://astral.sh/uv/install.sh | sh
Gates applied: no_behavioural_pass.
7168561a9c98full audit observations/trust-audit/mcp-server/pathintegral-institute__mcp-science.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-07 | 7168561a9c98 | CAUTION | B | 86 | first audit |
Questions
What is the MCP.science MCP server?
Open Source MCP Servers for Scientific Research
What tools does MCP.science expose?
42 in total: 24 read-only, 14 that write, and 4 that can delete or overwrite (delete_documents, delete_quantum_system, drop_table, purge_table). Every one is listed on this page with its risk.
Is MCP.science safe to connect to an agent?
With care. The audit graded it B (86/100) and found 18 things worth knowing before you trust this server, listed below with the exact line each was found on. Separately from the audit: 4 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does MCP.science need?
It reads JINA_API_KEY, JUPYTER_SERVER_TOKEN, MP_API_KEY, NEMAD_API_KEY, SSH_PASSWORD, SSH_PRIVATE_KEY_FILE and TXYZ_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does MCP.science run?
It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on PyPI as web-fetch.
How current is this page?
The grade is for one exact copy of the source (7168561a9c98), read on 2026-10-07. The repository is watched and re-audited when it changes.