Can this skill track changes on competitor websites?

Yes, it includes built-in patterns for competitive monitoring, including snapshot diffing and a severity-based classification system for pricing or feature updates.

How does the workflow choose between different research tools?

It follows a specific decision tree: it starts with WebFetch for speed, upgrades to Tavily for enhanced markdown and semantic search, and falls back to a full agent-browser for JS-heavy or authenticated sites.

Do I need a Tavily API key to use this skill?

A Tavily API key is recommended for high-quality extraction but not required; the workflow automatically collapses to a 2-tier system (WebFetch and agent-browser) if the key is missing.

Does it work with modern web frameworks like React or Vue?

Yes, it includes specific SPA detection patterns and uses the agent-browser fallback to wait for network idle and render dynamic content correctly.

Is this skill suitable for extracting technical documentation?

Absolutely. It is optimized to extract raw markdown from documentation sites, making it an ideal tool for the 'documentation-specialist' agent role.

Web Research & Monitoring Workflow

Name: Web Research & Monitoring Workflow
Author: yonatangross

byyonatangross

•

Web Scraping & Data Collection

Automates intelligent web content extraction and competitive intelligence gathering through a multi-tiered tool selection framework.

This skill provides a robust framework for Claude to perform high-quality web research by automatically selecting the most efficient tool—WebFetch, Tavily, or a full agent-browser—based on site complexity and technical requirements. Beyond simple scraping, it excels at competitive intelligence by monitoring target sites for changes in pricing or features, providing structured snapshot diffing, and classifying the severity of updates. It is designed to handle everything from simple static pages to complex Single Page Applications (SPAs) and authenticated sessions, ensuring clean markdown extraction for any research task.

Key Features

01Seamless handling of JS-rendered content, SPAs, and multi-page crawling patterns

02High-fidelity markdown extraction optimized for LLM context and documentation

03Configurable alerting for price increases, feature removals, or positioning shifts

04Intelligent 3-tier tool selection (WebFetch, Tavily, or agent-browser) based on site characteristics

05Competitive monitoring with automated snapshot diffing and change classification

0692 GitHub stars

Use Cases

01Automating competitive intelligence by tracking pricing and feature updates across market rivals

02Deep-dive web research for product strategy, market analysis, and technical discovery

03Extracting clean documentation from developer portals to provide technical context for coding tasks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yonatangross/orchestkit web-research-workflow

For use in Claude.ai and ChatGPT

Download Skill