active-research
Deep research and analysis tool. Generates comprehensive HTML reports on any topic, domain, paper, or technology. Enhanced with advanced browser automation — SPA handling, network idle wait, batch operations, stealth browsing, and intelligent page analysis. Use when user asks to research, analyze, investigate, deep-dive, or generate a report on any subject.
Security Assessment
About active-research
The active-research skill performs deep research on any topic, domain, paper, or technology and generates a comprehensive HTML report. It is built around the Actionbook Browser CLI, giving it robust web access for modern, dynamic sites where simple fetching falls short.
It uses enhanced browser capabilities including network-idle waiting for single-page apps, waiting on JavaScript conditions, focused page snapshots with token budgets, automatic dialog dismissal, image blocking for speed, stealth and anti-detection browsing, batch multi-step actions, and one-shot fetch. It detects topic type (arXiv papers via arXiv IDs or DOIs, academic keywords, URLs, or general text) and routes to arXiv advanced search or web search accordingly, then synthesizes the findings into a JSON report that renders to HTML. The skill mandates using actionbook browser commands for all web access rather than curl, scripting HTTP libraries, or generic fetch tools.
It is most useful when a user wants to research, investigate, or deep-dive a subject and produce a polished report, especially for content behind single-page apps, dynamic loading, or anti-bot protection.
FAQ
What does it produce?
A comprehensive research report (JSON that renders to a styled HTML report) on the requested topic, paper, or technology.
How does it access the web?
Exclusively through the Actionbook Browser CLI (actionbook browser, search, and get), not curl, scripting HTTP libraries, or generic fetch tools.
What browser capabilities does it use?
Network-idle waiting for single-page apps, JavaScript-condition waits, focused snapshots, auto-dismiss dialogs, image blocking, stealth and fingerprint rotation, batch actions, and one-shot fetch.
How does it handle academic topics?
It detects arXiv IDs, DOIs, and academic keywords and routes to arXiv advanced search with deep reads, supplementing with web search for non-academic sources.
Why use it over a simple fetch?
It reliably extracts content from single-page apps, dynamically loaded pages, and anti-bot-protected sites where simple HTTP fetching returns empty or blocked results.
Install active-research
Quick Setup:
- Copy the skill folder to
.claude/skills/ - Claude will automatically detect and use the skill
Repository
actionbook/actionbook