What types of selectors does it provide?

It returns both CSS and XPath selectors along with metadata about the element type and allowed interaction methods like click, fill, or type.

Can I search for specific web actions?

Yes, you can use the search_actions tool with keywords like 'login' or 'checkout' to find pre-computed manuals for specific sites.

Is this skill specific to a particular browser tool?

No, it provides standardized selectors and metadata that can be used with any browser automation library such as Playwright, Puppeteer, or Selenium.

How does Core Actionbook save tokens?

It provides specific, pre-mapped element selectors rather than requiring the AI to process and parse large amounts of raw HTML data.

Does it support hybrid search?

Yes, it supports vector, full-text, and hybrid search methods to ensure you find the most relevant action manuals for your query.

Core Actionbook

Name: Core Actionbook
Author: lywa1998

bylywa1998

Browser Automation

Simplifies browser automation by providing pre-computed element selectors and structured page interaction manuals.

About

Core Actionbook is a specialized skill designed to optimize browser-based AI workflows by replacing expensive HTML parsing with a registry of pre-computed action manuals. It provides agents with structured page information, including precise CSS/XPath selectors and allowed interaction methods (click, type, extract), which drastically reduces token consumption and improves the reliability of web automation. By leveraging hybrid search capabilities, developers can quickly retrieve specific selectors for complex web tasks, making it an essential tool for building robust web scraping and automation agents.

Key Features

0 GitHub stars
Structured element metadata including interaction types and allowed methods
Reduced token overhead by bypassing raw DOM parsing
Instant retrieval of pre-computed CSS and XPath selectors
Detailed URL-based action IDs with content previews and relevance scores
Hybrid search (vector and full-text) for finding specific web actions

Use Cases

Automating complex multi-step web workflows like logins or form submissions
Enhancing the accuracy of AI agents performing browser-based tasks
Building reliable web scraping scripts without manual selector identification

About

Key Features

0 GitHub stars
Structured element metadata including interaction types and allowed methods
Reduced token overhead by bypassing raw DOM parsing
Instant retrieval of pre-computed CSS and XPath selectors
Detailed URL-based action IDs with content previews and relevance scores
Hybrid search (vector and full-text) for finding specific web actions

Use Cases

Automating complex multi-step web workflows like logins or form submissions
Enhancing the accuracy of AI agents performing browser-based tasks
Building reliable web scraping scripts without manual selector identification