wisp-science

Open-source, local-first desktop AI research workbench for scientific computing with Python/R, MCP bioinformatics tools, SSH/WSL/GPU runtimes, and OpenAI/Anthro

What is wisp-science?

Open-source, local-first desktop AI research workbench for scientific computing with Python/R, MCP bioinformatics tools, SSH/WSL/GPU runtimes, and OpenAI/Anthropic models.

Overview of wisp-science

What it is

Wisp Science (WISP: Workspace for Intelligent Scientific Practice) is an open-source, local-first desktop AI research workbench for scientific computing. It allows users to search literature, run Python and R code, query scientific databases, and keep records of figures, runs, decisions, and drafts in one project. Data, conversations, and credentials stay on the user's machines.

Key features

  • Supports OpenAI-compatible and Anthropic models, and can drive Codex / Claude Code over ACP. The agent reads and writes project files, runs shell, and loads reusable Skills without flooding the prompt. Approval gates stay on unless the user opts into Full Permission. * Provides persistent Python and R kernels that keep variables across cells and turns; each conversation gets its own isolated kernel so parallel sessions never share state. Users can register local, WSL, and SSH hosts, probe hardware, and submit long Runs with live logs. Keys are stored in the OS keyring, never in SQLite. * Includes access to PubMed, GEO, and approximately 80 other databases through bundled MCP servers. Offers offline previews for notebooks, PDFs, Office files, and images. Supports isolated explorations that do not touch the mainline, and a Publication Workspace that freezes manuscript revisions into verifiable Evidence Capsules. * Restart and the full history is restored. Users can undo file edits from a turn, attach artifacts, files, and runtimes with @, search saved sessions with #, and apply a skill with /. Encrypted manual sync and project transfer are available; nothing syncs in the background.

Use cases

Scientific computing and research where users need to query literature and databases, run Python/R code, and keep a record of all steps in a local-first environment. Suitable for bioinformatics, data analysis, and reproducible research workflows.

Getting started

  1. Open the download page, choose your operating system and processor, and download via Cloudflare. GitHub Releases is also available. 2. Open a bundled demo — no API key needed — to see a full RNA-seq trajectory. 3. Add a model in Settings → Models and start a project. Setup walkthrough: Quick Start · basic configuration · model profiles · ACP agents Build from source, CLI, and architecture: development.

License and pricing

The license is AGPL-3.0-only (except where a directory notes otherwise). The project is open-source and free to use. Pricing type is open_source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/xuzhougengwisp-science.svg)](https://yunodb.com/tools/xuzhougengwisp-science?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/xuzhougengwisp-science?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/xuzhougengwisp-science.svg" alt="Featured on YunoDB" height="20"></a>