Quick Setup

Follow these steps to get the researcher running with the default configuration.

1. Clone the repository:

   git clone https://github.com/Prateekray/Open-Deep-Researcher.git

2. Change into the project directory:

   cd Open-Deep-Researcher

3. Install Python dependencies with uv (installation guide: https://docs.astral.sh/uv/getting-started/installation/):

   uv sync

4. Copy the environment variable template (WSL command shown):

   cp .env.example .env

5. Open .env and fill in the required credentials:
   • TAVILY_API_KEY
   • GOOGLE_API_KEY (Gemini)
   • LANGSMITH_API_KEY
6. Start the LangGraph development server:

   uv run langgraph dev --allow-blocking

Docker Setup (Simple)

1. Install Docker Desktop: https://docs.docker.com/desktop/install/
2. Copy the sample env file:

   cp .env.example .env

3. Edit .env and add your API keys.
4. Create the persistent checkpoint volume (one-time):

   docker volume create langgraph-checkpoints

   • This must exist before docker compose up because the compose file declares an external volume.
   • Verify it exists: docker volume ls | grep langgraph-checkpoints
   • Inspect (optional): docker volume inspect langgraph-checkpoints
   • If you ever need to recreate it: stop containers, then docker volume rm langgraph-checkpoints and run the create command again.
5. Build and start the containers:

   docker compose up --build

6. Wait for the logs to settle, then open:
   • Backend API docs: http://127.0.0.1:2024
   • Agent Chat UI: http://127.0.0.1:3001
7. Launch LangGraph Studio at https://smith.langchain.com/studio/?baseUrl=http://127.0.0.1:2024.
8. When you are done, stop everything with:

   docker compose down