# actionbook/actionbook

Let your AI agent get the sources behind logins and paywalls.

Repository: https://github.com/actionbook/actionbook
Canonical: https://ross.abutalabs.com/products/actionbook
Homepage: https://actionbook.app
Language: Rust
License: Apache-2.0
License Family: permissive
Topics: ai-agents, browser-automation, rust, skills
Last push: 2026-08-12T06:59:11+00:00

## Health v2 (maintenance only)
Score: 74/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 97, release rhythm 77, longevity 17
- inputs: {"age_days": 248, "days_push": 21, "days_rel": 72, "gap_med": 0.0, "n_releases_24m": 83}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1584, forks 118 (observed 2026-08-28T04:05:07.410322+00:00)

## What it is
Actionbook is an AI agent that lives in a Chrome side panel and operates websites on your behalf, reading data behind logins and paywalls and delivering finished work like reports, CSVs, and drafts. It pairs a remote MCP connector with a Chrome extension, using action manuals to drive direct API calls with UI automation fallback, and supports saving repeatable workflows as Recipes.

## Use cases
- have an AI agent collect data from websites behind my logins
- automate filling forms and operating web apps I use daily
- scrape taglines from 200 company websites quickly
- extract data from paywalled or login-walled pages into a CSV
- run browser automation tasks in parallel across many tabs
- save a web workflow as a reusable recipe and rerun it with new inputs
- connect my AI coding agent to websites via MCP

## When to choose
- you want an AI agent that works inside your browser with your existing sessions and logins
- you need faster, more resilient web automation than snapshot-based agents on SPAs and virtual DOM sites
- you want to expose websites to AI agents through an MCP connector
- you need to run many web tasks concurrently across tabs

## When to avoid
- you need a headless or server-side scraping solution without a browser
- you only need simple HTTP API integrations with no UI automation
- you require fully open-source self-hosting without any hosted credits or accounts
- you need automation outside the Chrome browser, e.g. desktop apps or mobile

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, browser-extension, web-scraping, workflow-automation, mcp, developer-tools
- domain: browser-extensions, web-development, developer-tools
- platform: browser, rust, cross-platform
- tags: browser-automation, chrome-extension, mcp-connector, ai-browser-agent, web-agent, recipes, login-handling, side-panel-agent, ai-agents, automation

## Member repositories
- actionbook/actionbook (main) score 74

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:07.410322+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:55:56.609134+00:00, confidence not recorded.
  - readme: https://github.com/actionbook/actionbook (fetched 2026-08-28T04:05:07.410322+00:00, sha 68f4bf16bdec)
  - homepage: https://actionbook.app (fetched 2026-08-29T11:26:21.213920+00:00, sha 40079c5dabb3)
  - site_page: https://actionbook.app/pricing (fetched 2026-08-29T11:26:21.216636+00:00, sha 70edbb2572a5)
- Data as of 2026-08-30T08:39:29.467469+00:00.