jobsonlook/xhs-mcp

小红书MCP服务 x-s x-t js逆向

What it solves

This project provides a way for AI models to interact with Xiaohongshu (Little Red Book), allowing them to search for notes, read content, and manage comments without requiring a heavy browser automation tool like Playwright.

How it works

It implements the Model Context Protocol (MCP) server, which allows AI clients (like Claude Desktop) to call specific tools. The server uses JavaScript reverse engineering to handle the required x-s and x-t parameters for direct HTTP requests to the Xiaohongshu API, requiring only a user's session cookie for authentication.

Who it’s for

Developers and AI users who want to integrate Xiaohongshu data retrieval and interaction capabilities into their LLM-powered agents or AI assistants.

Highlights

  • Direct API Access: Uses reverse-engineered headers to avoid the overhead of browser automation.
  • MCP Integration: Ready for use with MCP-compatible clients like Claude Desktop.
  • Comprehensive Toolset: Includes tools for searching notes, fetching home feed recommendations, reading note content and comments, and posting comments.

Related

  • Project
    browserbase/stagehand-python

    An AI-powered browser automation framework that combines natural language instructions with code to create flexible, self-healing web automations.

  • Project
    Rzna-5559/Mrite

    A one-stop repository for mathematical modeling competitions that integrates automated solving via Claude Code, LaTeX paper templates, and a library of common algorithms.

  • Project
    GoogleCloudPlatform/genai-for-marketing

    A Google Cloud-based solution that automates marketing content generation, trend analysis, and data querying using Vertex AI and Google Workspace integration.

  • Project
    agenvoy/Agenvoy

    A self-hosted AI agent harness that allows agents to automatically build, test, and reuse their own tools in a sandboxed environment.