Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/aipoch/open-science/llms.txt

Use this file to discover all available pages before exploring further.

Open Science is a standard Electron + npm project, so getting from zero to a running desktop app takes about five minutes. Follow the steps below to clone the repository, install dependencies, launch the app in development mode, and run your first AI-assisted analysis.

​
Prerequisites

Before you begin, make sure the following are available on your machine:
  • Node.js (LTS or newer) with npm included
  • Git
  • Python 3 (optional but recommended) — required if you want the built-in notebook kernel to execute code
1

Clone the repository

Clone the Open Science repository from GitHub and change into the project directory:
git clone https://github.com/aipoch/open-science.git
cd open-science
2

Install dependencies

Install all Node.js dependencies with npm:
npm install
The postinstall script runs automatically after npm install completes. It executes two setup steps back-to-back — prisma generate (which generates the local project database client) and electron-builder install-app-deps (which compiles native Electron dependencies for your platform). On a cold first run, this can take a few minutes. You do not need to run either command manually.
3

Start the app in development mode

Launch Open Science with hot-reload enabled:
npm run dev
This command builds the Electron main and preload bundles, starts the renderer dev server on localhost:5173, and opens the Open Science desktop window automatically. Development data is isolated under ~/.open-science-project, so it never touches a production install’s data (which lives under ~/.open-science).
Open the Chromium DevTools at any time by pressing F12 while the app is in focus. The DevTools console, network panel, and React developer tools are all available in dev mode — useful for inspecting IPC messages, renderer state, and agent activity logs.
Cold-start note: On the very first launch, the window occasionally fails to appear even though the process is running. If that happens, press Ctrl+C to stop the process and run npm run dev again.
4

Create your first project

Once the desktop window is open, you’ll land on the Home page.
  1. Click New Project in the sidebar or on the home screen.
  2. Enter a name for your project (for example, My First Analysis) and confirm.
  3. Open Science creates a new project workspace where all your sessions, artifacts, and notebook runs will be stored.
Each project is self-contained — sessions, agent conversations, notebook outputs, and generated files are all organized under the project you create here.
5

Start a conversation with the agent

With your project open, you’re ready to run your first AI-assisted analysis.
  1. Open a new Session inside your project.
  2. Type a task in the message input — for example:
    “Analyze this CSV file and summarize the key statistics for each numeric column.”
  3. Attach a file by clicking the attachment icon and selecting a CSV or other data file from your computer.
  4. Press Enter (or click Send) to submit the task.
The coordinating agent will plan the analysis, execute Python code in the built-in notebook kernel, store the output as artifacts, and display results — including rendered previews of CSV tables, plots, and Markdown summaries — directly in the conversation.The agent will ask for your approval before taking any higher-risk actions (such as accessing new data sources or making external requests). Approve or deny each action inline.

​
Next Steps

  • Read the Installation guide for pre-built binary downloads and platform-specific build instructions.
  • Browse ROADMAP.md in the repository to see the full capability map — what’s shipping today versus what’s planned next.
  • Join the Discord community to ask questions, share feedback, and follow the build in real time.

Build docs developers (and LLMs) love

Get started for freeTalk to us