WebHive

WebHive is a Python-based smart browser control center that lets you orchestrate AI-enabled browser tasks across multiple models and services while optionally using your own browser profiles. It supports running with a local server interface, access to an embedded UI, and integration with common AI providers like OpenAI and Anthropic through API keys. The server exposes a web-facing interface at the configured IP and port and can persist the browser session to maintain history between tasks when CHROME_PERSISTENT_SESSION is enabled. This makes it suitable for automation workflows, data collection, and browser-driven AI tasks where you want to preserve state between runs. Use the provided environment configuration to supply your API keys and browser-related settings, then start the server to begin issuing automated tasks or running AI agents inside a controlled browser context.

Prerequisites:

• Python 3.11+ installed on your system
• Git installed
• Optionally Docker if you prefer containerized usage

Local installation steps:

1. Clone the repository:

git clone git@github.com:webhiveos/WebHive.git
cd WebHive

2. Set up Python environment (uv recommended):

uv venv --python 3.11

Activate the environment:

• Windows (Command Prompt):

.venv\Scripts\activate

• Windows (PowerShell):

.\ .venv\Scripts\Activate.ps1

• macOS/Linux:

source .venv/bin/activate

3. Install Python dependencies:

uv pip install -r requirements.txt

4. Install Playwright (browser automation dependencies):

playwright install

5. Create and configure environment file:

cp .env.example .env

Edit .env to include your API keys and settings. 6. Run the WebHive server:

python webhive.py --ip 127.0.0.1 --port 7788

Docker installation (optional):

1. Ensure Docker and Docker Compose are installed.
2. Clone the repository and configure the .env file as above.
3. Build and start the container:

docker compose up --build

4. Access the interface at http://localhost:7788.

Tips and common considerations:

• Provide API keys for the LLM providers you intend to use (OPENAI_API_KEY, ANTHROPIC_API_KEY, GOOGLE_API_KEY).
• CHROME_PERSISTENT_SESSION controls whether the browser stays open between AI tasks; set to true if you want to preserve context, otherwise false for a fresh session each task.
• VNC_PASSWORD can be set in the .env file to secure VNC access; the default is optional but can be overridden.
• The default HTTP interface is bound to 127.0.0.1:7788; adjust --ip/--port if running behind a reverse proxy or on a different host.
• ARM64 builds are supported in Docker; ensure you pull the appropriate image for your architecture.
• If you encounter browser automation issues, ensure dependencies (e.g., Playwright) are installed and the environment is correctly activated.