Intelligent Internet Agent Explanation

Relevant source files

• README.md
• assets/ii.png
• pyproject.toml

This document provides comprehensive instructions for installing and configuring the II-Agent system on your local machine. It covers prerequisites, environment setup, installation steps, and initial configuration for both CLI and web interface usage.

For information about deployment in production environments, see Deployment.

Prerequisites

Before installing II-Agent, ensure your system meets the following requirements:

RequirementMinimum VersionNotes

Python	3.10+	Python 3.10 or higher is required
Node.js	18+	Only needed if using the frontend
Disk Space	1GB+	For dependencies and workspace files
Memory	4GB+ recommended	Higher is better for complex tasks
API Keys	Various	See Environment Configuration section

You'll also need one of the following:

• Anthropic API key
• Google Cloud project with Vertex AI API enabled

Sources: README.md94-98 pyproject.toml9

Installation Process

1. Clone the Repository

2. Set Up Python Environment

Create and activate a virtual environment:

3. Install Python Dependencies

Install the package in development mode:

This installs all required dependencies listed in the pyproject.toml file, including:

• Anthropic client libraries
• FastAPI and Uvicorn for WebSocket server
• Google Cloud AI Platform
• Browser automation tools (Playwright)
• PDF and media processing libraries
• Search tools integration

4. Frontend Setup (Optional)

If you plan to use the web interface, set up the frontend:

Sources: README.md133-147 pyproject.toml10-43

Environment Configuration

II-Agent requires several environment variables to function properly. These control API access, security settings, and integration points.

Environment Variables

Create a .env file in the root directory with the following variables:

Frontend Environment

If using the frontend, create another .env file in the frontend directory:

Sources: README.md100-131

Running II-Agent

II-Agent can be run in two primary modes:

1. Command Line Interface (CLI)
2. WebSocket Server with Web Interface

CLI Mode

The CLI provides direct interaction with the agent through your terminal.

Using Anthropic API:

Using Google Vertex AI:

CLI Options:

OptionDescriptionDefault

--project-id	Google Cloud project ID	None
--region	Google Cloud region	None
--workspace	Path to workspace directory	./workspace
--needs-permission	Require permission before executing commands	False
--minimize-stdout-logs	Reduce the amount of logs printed	False

Web Interface

To use the web interface, you need to:

1. Start the WebSocket server:

2. Start the frontend (in a separate terminal):

3. Access the interface at http://localhost:3000

Sources: README.md149-193

Workspace and Data Management

II-Agent organizes its files and data in a workspace directory, which maintains state between sessions.

The workspace directory contains:

• Saved files from agent operations
• Downloaded web content
• Generated media
• Session-specific context

You can specify a custom workspace location with the --workspace option when starting the agent.

Sources: README.md166

Troubleshooting

API Key Issues

If you encounter authentication errors:

• Verify your API keys in the .env file
• Ensure .env file is in the correct location (project root)
• Check that your Anthropic or Google Cloud credentials are valid

Dependency Problems

If you experience module import errors:

• Ensure you're using Python 3.10+
• Verify that the virtual environment is activated
• Try reinstalling dependencies: pip install -e .

Frontend Connection Issues

If the frontend cannot connect to the backend:

• Check that the WebSocket server is running
• Verify that STATIC_FILE_BASE_URL matches the WebSocket server address
• Ensure NEXT_PUBLIC_API_URL is correctly set in the frontend .env

Workspace Permissions

If you encounter file permission errors:

• Check that your user account has write access to the workspace directory
• Try running with elevated permissions if necessary

Sources: README.md100-131 README.md149-193

Next Steps

After successfully installing and configuring II-Agent, see:

• Usage Examples for practical examples of common tasks
• Command Line Interface for detailed CLI documentation
• WebSocket Server for API details
• Tool System to learn about available capabilities