explaingit

aipoch/open-science

Analysis updated 2026-05-18

334TypeScriptAudience · researcherComplexity · 3/5LicenseSetup · easy

TLDR

Open Science is a free desktop app where you describe a research task in plain language and an AI agent reads files, runs code, and produces reports, tables, and figures.

Mindmap

mindmap
  root((open science))
    What it does
      Plain language research tasks
      AI agent runs code
      Generates reports and figures
      Inspectable activity history
    Tech stack
      TypeScript
      Electron desktop app
    Use cases
      Data analysis
      Scientific reports
      Research skills catalog
    Audience
      Researchers
      Scientists

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Describe a data analysis task in plain language and let the agent read files, run code, and generate a report.

USE CASE 2

Attach scientific datasets and ask the agent to produce tables and figures for a research project.

USE CASE 3

Use built-in research skills and data connectors to speed up recurring scientific workflows.

What is it built with?

TypeScriptElectron

How does it compare?

aipoch/open-scienceinbrainfun/inbrainsignal-trade-core/weather-prediction-bot
Stars334332336
LanguageTypeScriptTypeScriptTypeScript
Setup difficultyeasymoderatehard
Complexity3/54/54/5
Audienceresearcherdeveloperdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · easy Time to first run · 30min

Download an installer for macOS, Windows, or Linux and connect a model provider during first-time setup.

Permissive Apache 2.0 license, use freely for any purpose including commercial use, with patent protection.

In plain English

Open Science is a free desktop application that gives researchers an AI assistant for scientific work. Instead of writing code or scripts yourself, you describe a task in plain language, such as analyzing a dataset or producing a report, and an AI agent inside the app reads your files, runs code, searches the web, calls scientific data connectors, and returns reports, tables, and figures, along with a record of every step it took. The app works as a project based workspace. You create a project, open a session, describe your goal along with the input data and any constraints, attach the relevant files, pick a verified AI model, and choose how much you want to approve each action the agent takes before it runs. As the agent works, you can inspect exactly which tools it used, which files it read or edited, and which searches or connectors it called, and you can open the generated results directly in a preview panel inside the app. Open Science is described as model agnostic, meaning you are not locked into one AI provider. During first time setup you can connect a built in model provider, an Anthropic compatible custom gateway, or reuse an existing subscription login such as a Claude Code login or a ChatGPT or Codex subscription, without needing to enter a separate API key. The app can also install its own managed runtime automatically if needed, without requiring you to separately install Node.js or have administrator access. The project includes previews for common scientific file types such as data tables, documents, images, source code, molecular structures and reactions, and notebook history, and it ships with a growing catalog of file based research skills that can be extended with custom skill packages. Installers are available for macOS, Windows, and Linux from the project's releases page, and the project is released under the Apache 2.0 license.

Copy-paste prompts

Prompt 1
How do I set up Open Science for the first time and connect a model provider like Claude Code or ChatGPT?
Prompt 2
Explain the plan, execute, produce, preview workflow that Open Science follows for a research task.
Prompt 3
What kinds of scientific file previews does Open Science support, like molecular structures or notebooks?
Prompt 4
How do I create a custom research skill package for Open Science?

Frequently asked questions

What is open-science?

Open Science is a free desktop app where you describe a research task in plain language and an AI agent reads files, runs code, and produces reports, tables, and figures.

What language is open-science written in?

Mainly TypeScript. The stack also includes TypeScript, Electron.

What license does open-science use?

Permissive Apache 2.0 license, use freely for any purpose including commercial use, with patent protection.

How hard is open-science to set up?

Setup difficulty is rated easy, with roughly 30min to a first successful run.

Who is open-science for?

Mainly researcher.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.