TwitterAPI MCP Server
An MCP server that provides access to Twitter data and write actions through TwitterAPI.io, bypassing the need for a standard Twitter developer account. It enables users to search tweets, retrieve profiles, and post content with support for pagination and enterprise proxy configurations.
README
TwitterAPI MCP Server
A Model Context Protocol (MCP) server that provides access to Twitter data through the TwitterAPI.io service. This server enables Claude and other MCP clients to interact with Twitter's ecosystem without requiring Twitter developer account approval.
Attribution: This project is a fork of kinhunt/twitterapi-mcp with bug fixes and improvements to match the official TwitterAPI.io documentation.
Features
- User Information: Get detailed user profiles, followers, and following lists
- Tweet Operations: Search tweets, get tweet details, replies, and user timelines
- Search Capabilities: Advanced search for both tweets and users
- Write Actions: Post tweets and interact with content (requires login)
- Pagination Support: All list endpoints support cursor-based pagination
- Enterprise Ready: Proxy support and robust error handling
- No Twitter Auth: Uses TwitterAPI.io which doesn't require Twitter developer approval
Installation
Quick Start with npx (Recommended)
npx twitterapi-mcp-server
Global Installation
npm install -g twitterapi-mcp-server
Local Installation
npm install twitterapi-mcp-server
Configuration
Getting an API Key
- Visit TwitterAPI.io
- Create an account and log in
- Get your API key from the dashboard
- The API key format looks like:
new1_xxxxxxxxxxxxxxxxxxxxx
Environment Variables
| Variable | Required | Description |
|---|---|---|
TWITTERAPI_API_KEY |
Yes | Your TwitterAPI.io API key |
PROXY_URL |
No | Proxy URL for enterprise environments |
HTTP_PROXY |
No | Alternative proxy configuration |
HTTPS_PROXY |
No | Alternative proxy configuration |
MCP Client Configuration
Claude Desktop
Add this to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"twitterapi": {
"command": "npx",
"args": ["-y", "twitterapi-mcp-server"],
"env": {
"TWITTERAPI_API_KEY": "your_api_key_here"
}
}
}
}
Claude Desktop with Proxy
{
"mcpServers": {
"twitterapi": {
"command": "npx",
"args": ["-y", "twitterapi-mcp-server"],
"env": {
"TWITTERAPI_API_KEY": "your_api_key_here",
"PROXY_URL": "http://proxy.company.com:8080"
}
}
}
}
Cursor IDE
Add to your Cursor MCP settings (.cursor/mcp.json):
{
"mcpServers": {
"twitterapi": {
"command": "npx",
"args": ["-y", "twitterapi-mcp-server"],
"env": {
"TWITTERAPI_API_KEY": "your_api_key_here"
}
}
}
}
Claude Code
Add to your Claude Code MCP settings (~/.claude/settings.json):
{
"mcpServers": {
"twitterapi": {
"command": "npx",
"args": ["-y", "twitterapi-mcp-server"],
"env": {
"TWITTERAPI_API_KEY": "your_api_key_here"
}
}
}
}
Using with Node directly
If you prefer to run with Node directly instead of npx:
{
"mcpServers": {
"twitterapi": {
"command": "node",
"args": ["/path/to/twitterapi-mcp-server/build/index.js"],
"env": {
"TWITTERAPI_API_KEY": "your_api_key_here"
}
}
}
}
Available Tools
User Information
| Tool | Description | Required Params | Optional Params |
|---|---|---|---|
get_user_by_username |
Get user details by username | username |
- |
get_user_by_id |
Get user details by user ID | user_id |
- |
get_user_followers |
Get user's followers (200/page) | username |
cursor, pageSize |
get_user_following |
Get users someone follows (200/page) | username |
cursor, pageSize |
search_users |
Search for users by keyword | query |
cursor |
Tweet Operations
| Tool | Description | Required Params | Optional Params |
|---|---|---|---|
get_user_tweets |
Get tweets from a user (20/page) | username or userId |
cursor, includeReplies |
search_tweets |
Search tweets by keywords | query |
queryType (Latest/Top), cursor |
get_tweet_by_id |
Get tweets by IDs | tweet_ids (array) |
- |
get_tweet_replies |
Get replies to a tweet (20/page) | tweetId |
cursor, sinceTime, untilTime |
Write Actions (Requires Login)
| Tool | Description | Required Params | Optional Params |
|---|---|---|---|
login_user |
Login to Twitter account | user_name, email, password, proxy |
totp_secret |
create_tweet |
Post new tweets | tweet_text, proxy |
reply_to_tweet_id, attachment_url, media_ids |
Examples
Get User Information
// Get user by username
await get_user_by_username({ username: "elonmusk" })
// Get user followers with pagination
await get_user_followers({
username: "elonmusk",
pageSize: 100
})
// Get next page using cursor
await get_user_followers({
username: "elonmusk",
cursor: "next_cursor_from_previous_response"
})
Search and Retrieve Tweets
// Search latest tweets
await search_tweets({
query: "artificial intelligence",
queryType: "Latest"
})
// Search top/popular tweets
await search_tweets({
query: "OpenAI",
queryType: "Top"
})
// Advanced search with operators
await search_tweets({
query: "AI from:elonmusk since:2024-01-01"
})
// Get user's recent tweets
await get_user_tweets({ username: "openai" })
// Get user's tweets including replies
await get_user_tweets({
username: "openai",
includeReplies: true
})
// Get specific tweets by IDs
await get_tweet_by_id({
tweet_ids: ["1234567890123456789", "9876543210987654321"]
})
// Get replies to a tweet
await get_tweet_replies({
tweetId: "1234567890123456789"
})
Create Content (Requires Login)
// Login first (requires residential proxy)
await login_user({
user_name: "your_username",
email: "your_email@example.com",
password: "your_password",
proxy: "http://user:pass@proxy:port"
})
// Post a tweet
await create_tweet({
tweet_text: "Hello from MCP!",
proxy: "http://user:pass@proxy:port"
})
// Reply to a tweet
await create_tweet({
tweet_text: "Great point!",
reply_to_tweet_id: "1234567890123456789",
proxy: "http://user:pass@proxy:port"
})
Pagination
All list endpoints return paginated results with cursor-based navigation:
{
"data": [...],
"has_next_page": true,
"next_cursor": "cursor_string_for_next_page"
}
To get the next page, pass the next_cursor value as the cursor parameter in your next request.
API Pricing
TwitterAPI.io offers pay-as-you-go pricing:
| Operation | Price |
|---|---|
| Tweets | $0.15 per 1,000 |
| User profiles | $0.18 per 1,000 |
| Followers/Following | $0.15 per 1,000 |
| Login | $0.003 per call |
| Create tweet | $0.003 per call |
- Minimum charge: $0.00015 per request
- No monthly fees
- Free trial credits available
- Discounted rates for students and research institutions
Development
Building from Source
git clone https://github.com/Jing-yilin/twitterapi-mcp-server.git
cd twitterapi-mcp-server
npm install
npm run build
Running Tests
# Using bun
bun test
# Or with npm (requires bun installed)
npm test
Testing the Server Manually
# Set your API key
export TWITTERAPI_API_KEY="your_api_key"
# Test tools list
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/list", "params": {}}' | node build/index.js
# Test a tool call
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "get_user_by_username", "arguments": {"username": "elonmusk"}}}' | node build/index.js
Project Structure
twitterapi-mcp-server/
├── src/
│ ├── index.ts # Main server implementation
│ ├── index.test.ts # Unit tests
│ └── integration.test.ts # Integration tests
├── build/ # Compiled JavaScript
├── package.json
├── tsconfig.json
└── README.md
Error Handling
The server handles common errors:
| Error | Cause | Solution |
|---|---|---|
| 401 Unauthorized | Invalid API key | Check your TWITTERAPI_API_KEY |
| 402 Payment Required | Insufficient credits | Add credits at TwitterAPI.io dashboard |
| 429 Rate Limited | Too many requests | Wait and retry, or reduce request rate |
| 400 Bad Request | Invalid parameters | Check parameter names and formats |
Security Considerations
- Store API keys as environment variables, never in code
- Login credentials are used only for authentication, not stored persistently
- All API requests use HTTPS
- Proxy support available for enterprise security requirements
- The
login_cookiefrom login is stored in memory only for the session
Troubleshooting
"Unauthorized" error
- Verify your API key is correct
- Check that
TWITTERAPI_API_KEYenvironment variable is set
"Credits not enough" error
- Add credits to your TwitterAPI.io account
- Check your usage at the dashboard
Server not starting
- Ensure Node.js >= 18.0.0 is installed
- Run
npm run buildto compile TypeScript - Check for error messages in stderr
Proxy issues
- Verify proxy URL format:
http://user:pass@host:port - Test proxy connectivity independently
- For HTTPS proxies, use
HTTPS_PROXYvariable
Contributing
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
License
This project is licensed under the MIT License - see the LICENSE file for details.
Support
- API Documentation: docs.twitterapi.io
- Issues: GitHub Issues
- TwitterAPI.io Support: twitterapi.io
Acknowledgments
- Originally forked from kinhunt/twitterapi-mcp
- Built on TwitterAPI.io service
- Uses the Model Context Protocol
- Part of the growing MCP ecosystem
Note: This is an unofficial MCP server for TwitterAPI.io. Make sure to comply with Twitter's Terms of Service when using this tool.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。