Skip to content

Browser action engine for AI agents. 10× faster, resilient by design.

License

Notifications You must be signed in to change notification settings

CatchNip/actionbook

 
 

Repository files navigation

Actionbook Cover

Actionbook

GitHub last commit NPM Downloads npm version skills

Browser Action Engine for AI Agents
Actionbook provides up-to-date action manuals and DOM structure,
so your agent operates any website instantly without guessing.

Website · GitHub · X · Discord


Table of Contents

Why Actionbook?

❌ Without Actionbook

Building reliable browser agents is difficult and expensive:

  • Slow Execution: Agents waste time parsing full HTML pages to find elements.
  • High Token Costs: Sending entire DOM trees to LLMs consumes massive context windows.
  • Brittle Selectors: Updates to website UIs break hardcoded selectors and agent logic immediately.
  • Hallucinations: LLMs often guess incorrect actions when faced with complex, unstructured DOMs.

✅ With Actionbook

Actionbook places up-to-date action manuals with the relevant DOM selectors directly into your LLM's context.

  • 10x Faster: Agents access pre-computed "Action manuals" to know exactly what to do without exploring.
  • 100x Token Savings: Instead of whole HTML page, agents receive only related DOM elements in concise, semantic JSON definitions.
  • Resilient Automation: Action manuals are maintained and versioned. If a site changes, the manual is updated, not your agent.
  • Universal Compatibility: Works with any LLM (OpenAI, Anthropic, Gemini) and any AI operator framework.

See how Actionbook enables an agent to complete an Airbnb search task 10x faster.

actionbook-demo.mp4

Quick Start

Get started with Actionbook in under 2 minutes:

Step 1: Install the CLI

npm install -g @actionbookdev/cli

The Rust-based CLI uses your existing system browser (Chrome, Brave, Edge, Arc, Chromium), so no extra browser install step is required.

Step 2: Use with any AI Agent

When working with any AI coding assistant (Claude Code, Cursor, etc.), add this to your prompt:

Use Actionbook to understand and operate the web page.

The agent will automatically use the CLI to fetch action manuals and execute browser operations.

Step 3 (Optional): Add the Skill

For enhanced agent integration, add the Actionbook skill:

npx skills add actionbook/actionbook

Installation

# Install the Actionbook CLI globally
npm install -g @actionbookdev/cli

# Initialize local configuration
actionbook setup

The CLI is all you need to get started. For advanced use cases, Actionbook also offers an MCP Server and JavaScript SDK.

Examples

Explore real-world examples in the Examples Documentation.

Available Tools

Actionbook provides tools for searching and retrieving action manuals. See the CLI Reference for the full command list. If you're using the MCP integration, see the MCP Tools Reference.

Documentation

For comprehensive guides, API references, and tutorials, visit our documentation site:

actionbook.dev/docs

Stay tuned

We move fast. Star Actionbook on Github to support and get latest information.

Star Actionbook

Join the community:

Development

This is a monorepo using pnpm workspaces and Turborepo.

Prerequisites

  • Node.js >= 18 (20+ recommended)
  • pnpm >= 10
  • PostgreSQL database (local or hosted like Neon / Supabase)

First-time Setup

  1. Install dependencies:
pnpm install
  1. Configure environment variables by copying .env.example to .env in the following packages:

    • services/db
    • apps/api-service
    • services/action-builder (optional, for recording)
    • services/knowledge-builder (optional, for knowledge extraction)
  2. Run database migrations:

cd services/db && pnpm migrate

Start the Development Server

pnpm dev

Contributing

  • Request a Website - Suggest websites you want Actionbook to index.
  • Join the Waitlist - We are currently in private beta. Join if you are interested in contributing or using Actionbook.

License

See LICENSE for the license details.

About

Browser action engine for AI agents. 10× faster, resilient by design.

Resources

License

Code of conduct

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • TypeScript 55.0%
  • Rust 31.5%
  • JavaScript 9.9%
  • Shell 1.1%
  • MDX 1.0%
  • Python 0.8%
  • Other 0.7%