X.com MCP Server

X.com MCP Server

Model Context Protocol server that enables LLMs to interact with X.com (formerly Twitter) through OAuth 2.0 authentication, supporting major Post-related operations including reading, writing, searching, and managing posts, likes, retweets, and bookmarks.

Category
访问服务器

README

X.com MCP Server

A Model Context Protocol server that provides access to X.com's API capabilities. This server enables LLMs to interact with X.com (formerly Twitter) through OAuth 2.0 authentication, supporting all major Post-related operations including reading, writing, searching, and managing posts, likes, retweets, and bookmarks.

Features

  • OAuth 2.0 Authentication - Secure Bearer token authentication
  • Complete Post API Coverage - All X.com Post endpoints from the official API
  • Type Safety - Full TypeScript implementation with Zod validation
  • Comprehensive Tools - 21 tools covering all major X.com operations

Available Tools

Lookup

  • getSinglePost - Retrieve a single post by ID with optional field expansions
  • getBulkPosts - Retrieve up to 100 posts by their IDs

Manage Posts

  • createPost - Create a new post with text, media, polls, replies, etc.
  • deletePost - Delete a post by ID
  • hideReply - Hide or unhide a reply to a post

Timelines

  • getUserTimeline - Get a user's timeline of posts
  • getUserMentions - Get posts that mention a specific user

Search

  • searchRecent - Search recent posts (last 7 days)
  • searchAll - Search all posts (full archive - requires Academic/Enterprise access)

Post Counts

  • getPostCountsRecent - Get time-bucketed post counts for recent posts
  • getPostCountsAll - Get time-bucketed post counts for all posts (Academic/Enterprise)

Retweets

  • getRetweets - Get posts that retweet a specific post
  • createRetweet - Retweet a post
  • deleteRetweet - Remove a retweet

Likes

  • getLikingUsers - Get users who liked a specific post
  • getLikedTweets - Get posts that a user has liked
  • likePost - Like a post
  • unlikePost - Unlike a post

Bookmarks

  • getUserBookmarks - Get a user's bookmarked posts
  • bookmarkPost - Bookmark a post
  • removeBookmark - Remove a bookmark

Installation

Using pnpm (recommended)

pnpm install
pnpm run build

After installation, you can run it using:

node dist/index.js

Using Docker

docker build -t x.com-mcp .
docker run -i --rm -e X_COM_ACCESS_TOKEN=your-access-token x.com-mcp

Configuration

Environment Variables

  • X_COM_ACCESS_TOKEN (required): Your X.com OAuth 2.0 access token

Authentication Setup

  1. Create an X.com Developer Account at developer.x.com
  2. Create a new project and app
  3. Generate OAuth 2.0 credentials
  4. Implement the OAuth 2.0 Authorization Code with PKCE flow
  5. Store the resulting access token as X_COM_ACCESS_TOKEN

Configure for Claude.app

Add to your Claude settings:

<details> <summary>Using pnpm</summary>

"mcpServers": {
  "x-com": {
    "command": "node",
    "args": ["dist/index.js"],
    "env": {
      "X_COM_ACCESS_TOKEN": "your-access-token"
    }
  }
}

</details>

<details> <summary>Using Docker</summary>

"mcpServers": {
  "x-com": {
    "command": "docker",
    "args": ["run", "-i", "--rm", "-e", "X_COM_ACCESS_TOKEN=your-access-token", "x.com-mcp"]
  }
}

</details>

Example Interactions

  1. Create a post:
{
  "name": "createPost",
  "arguments": {
    "text": "Hello world! This is my first post via the MCP server.",
    "reply_settings": "everyone"
  }
}
  1. Search recent posts:
{
  "name": "searchRecent",
  "arguments": {
    "query": "artificial intelligence",
    "max_results": 10,
    "tweet.fields": "created_at,author_id,public_metrics"
  }
}
  1. Get user timeline:
{
  "name": "getUserTimeline",
  "arguments": {
    "id": "123456789",
    "max_results": 20,
    "expansions": "author_id"
  }
}
  1. Like a post:
{
  "name": "likePost",
  "arguments": {
    "user_id": "123456789",
    "tweet_id": "987654321"
  }
}
  1. Get post counts:
{
  "name": "getPostCountsRecent",
  "arguments": {
    "query": "machine learning",
    "granularity": "day"
  }
}

API Endpoints Covered

This server implements all Post-related endpoints from the X.com API v2 specification:

  • Lookup: /2/tweets/:id, /2/tweets
  • Manage Posts: /2/tweets (POST), /2/tweets/:id (DELETE), /2/tweets/:id/hidden (PUT)
  • Timelines: /2/users/:id/tweets, /2/users/:id/mentions
  • Search: /2/tweets/search/recent, /2/tweets/search/all
  • Post Counts: /2/tweets/counts/recent, /2/tweets/counts/all
  • Retweets: /2/tweets/:id/retweets, /2/users/:id/retweets
  • Likes: /2/tweets/:id/liking_users, /2/users/:id/liked_tweets, /2/users/:id/likes
  • Bookmarks: /2/users/:id/bookmarks

Required Scopes

Ensure your X.com app has the following OAuth 2.0 scopes:

  • tweet.read - Read posts and user information
  • tweet.write - Create and delete posts
  • tweet.moderate.write - Hide/unhide replies
  • users.read - Read user information
  • like.read - Read likes information
  • like.write - Like and unlike posts
  • bookmark.read - Read bookmarks
  • bookmark.write - Create and delete bookmarks

Debugging

You can use the MCP inspector to debug the server:

npx @modelcontextprotocol/inspector node dist/index.js

Examples of Questions for Claude

  1. "Can you search for recent posts about artificial intelligence?"
  2. "Create a post about the benefits of TypeScript."
  3. "Find posts that mention machine learning and show their engagement metrics."
  4. "Get my recent bookmarks and analyze the topics."

Build

# Install dependencies
pnpm install

# Build TypeScript
pnpm run build

# Start server
pnpm start

Rate Limits

Be aware of X.com API rate limits:

  • Most read operations: 300 requests per 15 minutes
  • Write operations: 50 requests per 15 minutes
  • Bookmark operations: 50 write / 180 read per 15 minutes

Refer to the X.com API documentation for detailed rate limit information.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选