We just open-sourced a server that implements the Model Context Protocol (MCP), designed to connect LLMs and AI apps (like Claude Desktop, Cursor, and Windsurf) directly to Decodo’s platform — no glue code required. This makes it easy for agents and copilots to access the web, retrieve structured data, and run real scraping tasks through a common protocol. The server includes built-in tools for things like: Scraping websites as Markdown Running Google or Amazon searches and returning parsed results Controlling geo-location, language, and token limits via prompt It’s privacy-conscious, region-flexible, and works out of the box with most MCP clients. You can run it locally, or install it in one click via Smithery. We’d love feedback from developers working on: LLM agents or autonomous tooling AI-based research or retrieval systems Scraping pipelines or data orchestration tools Here’s the repo: https://github.com/Decodo/decodo-mcp-server

Let us know what features you’d like to see added, or if you run into any integration issues. We're actively improving this based on early use cases and real-world friction. — The Decodo team