actionbook

Actionbook

Actionbook turns the websites you work in every day into something your AI agent can actually operate. Direct API requests when possible, UI automation when not, with login handled. Fast and resilient.

Table of Contents

• Why Actionbook?
• Installation
• Quick Start
• AI Agent Skills
• Examples
• Available Tools
• Documentation
• Contributing

Why Actionbook?

❌ Without Actionbook

• Slow. Agents take a snapshot after every single step, parse the page, then decide what to do next. Searching one room on Airbnb takes 5 minutes.
• Brittle. Modern websites use virtual DOMs, streaming components, and SPAs. Agents don't understand these rendering mechanisms, so they fail on dropdowns, date pickers, and login walls.
• One at a time. Your agent finishes one page before it can start the next. Need to check 30 company websites? That's 30 rounds, one after another.

✅ With Actionbook

• 10x faster. Action manuals tell agents exactly what to do. No parsing, no guessing.
• Accurate. Direct API calls when possible, UI fallback when not. Login handled either way.
• Concurrent. Stateless architecture. Operate dozens of tabs in parallel.

See an agent visits 192 First Round portfolio company websites and collects their taglines in 2 minutes. (Video is not sped up)

https://github.com/user-attachments/assets/35079a19-7236-47a8-87ed-3edf6436c2bf

Installation

Install via npm:

npm install -g @actionbookdev/cli

Or build from source:

cargo install --git https://github.com/actionbook/actionbook --path packages/cli --locked

The Rust-based CLI uses your existing system browser (Chrome, Brave, Edge, Arc, Chromium). Reuse your logged-in sessions for convenience, or launch a clean profile for privacy.

Quick Start

actionbook browser start

# Open tabs
actionbook browser open https://stripe.com --session s1
actionbook browser open https://linear.app --session s1
actionbook browser open https://vercel.com --session s1

# Operate all tabs concurrently
actionbook browser snapshot --session s1 --tab t1 &
actionbook browser snapshot --session s1 --tab t2 &
actionbook browser snapshot --session s1 --tab t3 &

# Interact with each tab using refs from its snapshot
actionbook browser click @e5 --session s1 --tab t1
actionbook browser fill @e3 "hello" --session s1 --tab t2
actionbook browser click @e8 --session s1 --tab t3

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.

AI Agent Skills

Actionbook ships with Agent Skills that teach your AI agent how to use the CLI.

Claude Code, Cursor, Codex, Windsurf, Antigravity, Opencode — one command:

npx skills add actionbook/actionbook

Hermes — one command (registers the skill in ~/.hermes/skills/):

hermes skills install skills-sh/actionbook/actionbook/actionbook -y

Then start a chat and say things like "use actionbook to open google.com and search for anthropic". Hermes auto-activates the skill and drives the CLI for you.

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.

Extension & Daemon