mcp-server

wisp-science: AI Research Workbench

Name: wisp-science: AI Research Workbench
Author: xuzhougeng

Get a local-first AI research workbench for scientific computing with Python/R.

1,102 stars115 forksRustGuide quality 8/10Updated 9/7/2026100% free · open source

What it does

Wisp-science is a local-first desktop AI research workbench for scientific computing that integrates Python, R, bioinformatics tools, and AI models from OpenAI and Anthropic.

When to use it

•When you need to run AI models locally for data privacy or security reasons
•When you require a desktop environment for scientific computing with Python, R, or bioinformatics tools
•When you want to utilize OpenAI or Anthropic models for research without relying on cloud services

Ready-to-paste prompt

Run a Python script with access to the workbench's tools: python your_script.py --mcp-tool your_mcp_tool

Heads up: Ensure you have the necessary dependencies, such as SSH, WSL, or GPU runtimes, installed and configured properly before starting the workbench, as their absence may hinder the workbench's functionality

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/wisp-science && curl -fsSL https://workflowstacks.com/api/skills/wisp-science/claude-skill -o ~/.claude/skills/wisp-science/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How wisp-science: AI Research Workbench works

Codeflow

Free to inspect

Wisp-science: AI Research Workbench is a very large Rust project (~347k lines across 479 code files, plus 17 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, AGPL-3.0 license, has a test suite.

Size

Very large codebase

~347k lines · 479 code files · days to read — use, don't read

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

Rust

No API keys detected

Rust 70%JavaScript 16%TypeScript 6%Python 5%CSS 3%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
AGENTS.md
The instructions the AI actually follows
3
Cargo.toml
Dependencies and the commands it exposes
4
skills/agent-infini/SKILL.md
Inside skills/ — an example of what the AI is told to do

What's in each folder

skills/Prompts, skills & agent definitions149 files

ui/Frontend / UI185 files

docs/Documentation114 files

crates/Folder256 files

src-tauri/Folder166 files

browser-extension/Folder20 files

ui-tests/Folder15 files

scripts/Helper scripts8 files

READMEHas testsDocumentedCI checksAGPL-3.0 licenseUpdated this month