Web Research
Summary
MCP-WebResearch provides AI assistants with web browsing capabilities for conducting research through Google searches and webpage content extraction. Built with TypeScript using Playwright for browser automation and Turndown for HTML-to-Markdown conversion, it offers tools to search Google, visit webpages, and capture screenshots while handling common challenges like consent dialogs and bot detection. The implementation includes an agentic research prompt that guides assistants through iterative research workflows, making it particularly valuable for deep topic exploration, fact-checking, and information gathering directly within AI conversations.
Available Actions(3)
search_google
Performs Google searches and extracts results. Arguments: { query: string }
visit_page
Visits a webpage and extracts its content. Arguments: { url: string, takeScreenshot?: boolean }
take_screenshot
Takes a screenshot of the current page. No arguments required
社区评论
暂无评论. 成为第一个评论的人！
登录以参与讨论
