Configure each repositories environment so the agent has the most optimal workspace to work with.

​
Repository Selection

Choose a repository to configure its environment settings. Select from your available repositories in the dropdown menu.

​
Image & Packages

​
Base Image

Specify the Docker image to use as the base environment. Common options include:
  • python:3.11
  • ubuntu:24.04
  • node:20

​
Package Management

Configure packages for your environment: APT Packages: Add system-level packages that need to be installed via apt-get PIP Packages: Add Python packages that should be installed via pip

​
Environment Variables & Secrets

Configure environment variables and secrets for your agent’s workspace:
  • Environment Variables: Stored in plain text and accessible to your agent
  • Secrets: Stored securely in the vault for sensitive information like API keys
You can add key-value pairs for both environment variables and secrets. Use the “Import .env” button to bulk import from an existing environment file.
Secrets are stored securely in a vault with encryption at rest. Once saved, secret values cannot be viewed for security reasons, but you can update them.Regular environment variables are stored in your environment settings and are visible in plain text.

​
Setup Script

Define custom setup commands that run after the environment is created. Each line will be executed separately in the same session. Example setup script: 
pip install -r requirements.txt
npm install
chmod +x scripts/setup.sh

​
Validate & Save Settings

Before finalizing your configuration, click the validate and save button. This initializes a sandbox environment with your configuration to ensure the commands work as expected. Once the validation finished, it will save automatically and every subsequent run in that repository will use your environment setup!
Don’t leave the page while validation is taking place, as the progress will be lost.