apify/rag-web-browser
Navigateur web pour les assistants OpenAI, les pipelines RAG ou les agents IA, similaire au navigateur web de ChatGPT. Il interroge Google Search, extrait les N meilleures pages et renvoie leur contenu au format Markdown pour un traitement ultérieur par un LLM. Il peut également extraire des URL individuelles. Prend en charge le mcp servers.
The RAG Web Browser is an Apify Actor designed to provide web browsing capabilities for AI agents, LLM applications, and RAG pipelines. Similar to ChatGPT's web browsing feature, it can query Google Search, scrape top N pages, clean HTML, convert content to Markdown, and return it for LLM processing. It also supports scraping individual URLs. This Actor functions as an MCP server via SSE, allowing seamless integration with tools like Claude Desktop and custom MCP clients. It offers performance optimizations for low-latency applications, including a standby web server mode and configurable scraping tools (raw-http for static sites, browser-playwright for dynamic sites). The project is open-source and developed by Apify.
Recommend MCP Servers 💡
@mario-andreschak/mcp_video_recognition
An MCP server for image, audio, and video recognition using Google's Gemini AI.
NCBI Datasets MCP
Model Context Protocol for the NCBI Datasets MCP server, following the MCP specification format.
whois-mcp
A WHOIS lookup service implemented as a Model Context Protocol (MCP) server
figma-developer-mcp
Provide Figma layout information to AI coding agents via MCP.
mcp-geo
An MCP service for retrieving user geolocation information and integrating it with large language models through the Model Context Protocol.
cursor-talk-to-figma-mcp
MCP integration between Cursor AI and Figma for reading and modifying designs programmatically
