Twitter MCP
MCP server for Twitter/X enabling AI agents to search, post, reply, and engage with tweets.
README
@isteam/twitter-mcp
MCP server for Twitter/X — search tweets, post, reply, and engage via AI agents.
Built by is.team — the AI-native project management platform.
Quick Start
Add to your MCP config (.mcp.json for Claude Code, or Claude Desktop settings):
{
"mcpServers": {
"twitter": {
"command": "npx",
"args": ["-y", "@isteam/twitter-mcp"],
"env": {
"TWITTER_API_KEY": "your-api-key",
"TWITTER_API_SECRET": "your-api-secret",
"TWITTER_ACCESS_TOKEN": "your-access-token",
"TWITTER_ACCESS_TOKEN_SECRET": "your-access-token-secret"
}
}
}
}
Tools (15)
Search & Discovery
| Tool | Description |
|---|---|
search_tweets |
Search tweets by keywords, hashtags, or phrases (max 100 results) |
get_mentions |
Get recent mentions of the authenticated user |
get_user_tweets |
Get a user's recent tweets by username |
Engagement
| Tool | Description |
|---|---|
post_tweet |
Post a new tweet (max 280 characters) |
reply_tweet |
Reply to a tweet |
quote_tweet |
Quote tweet with your commentary |
delete_tweet |
Delete a tweet |
like_tweet |
Like a tweet |
retweet |
Retweet a tweet |
follow_user |
Follow a user by username |
User & Profile
| Tool | Description |
|---|---|
get_me |
Get authenticated user info (followers, following, tweet count) |
get_tweet |
Get a specific tweet with author info and metrics |
get_user |
Look up a user by username |
update_profile |
Update your profile (name, bio, url, location) |
Environment Variables
| Variable | Required | Description |
|---|---|---|
TWITTER_API_KEY |
Yes | Twitter API key (consumer key) |
TWITTER_API_SECRET |
Yes | Twitter API secret (consumer secret) |
TWITTER_ACCESS_TOKEN |
Yes | OAuth 1.0a access token |
TWITTER_ACCESS_TOKEN_SECRET |
Yes | OAuth 1.0a access token secret |
Getting your credentials
- Go to the Twitter Developer Portal
- Create a project and app
- Set up OAuth 1.0a with read and write permissions
- Generate your access token and secret from the "Keys and tokens" tab
- Copy all four values into your MCP config
Rate Limits & Agent Safety
Twitter API v2 enforces per-endpoint rate limits in 15-minute windows:
| Endpoint | Limit | Notes |
|---|---|---|
POST /2/tweets (post, reply, quote) |
300 / 15 min | ~20/min |
GET /2/tweets/search/recent |
450 / 15 min | Requires Basic tier |
POST /2/users/:id/likes |
300 / 15 min | |
POST /2/users/:id/retweets |
300 / 15 min | |
POST /2/users/:id/following |
300 / 15 min | |
GET /2/users/:id/mentions |
450 / 15 min |
Idempotency note: Twitter API does not deduplicate identical tweets — posting the same text twice creates two tweets. If your agent retries on timeout, it may create duplicates. Use get_user_tweets to verify before retrying a post.
Backoff: When rate-limited, the API returns 429 Too Many Requests with a x-rate-limit-reset header (Unix timestamp). Wait until that time before retrying.
Usage Examples
Monitor your brand:
"Search for tweets mentioning 'is.team' in the last 24 hours and show engagement metrics"
Engage with your audience:
"Check my recent mentions and reply to any questions with helpful answers"
Post an update:
"Tweet: Just shipped our new AI-powered project management feature! Check it out at is.team"
About is.team
is.team is an AI-native project management platform where AI agents and humans collaborate as real teammates. AI agents join boards, create tasks, chat, and get work done — just like any other team member.
Part of the is.team open-source MCP ecosystem:
- @isteam/mcp — Project management
- @isteam/google-ads-mcp — Google Ads
- @isteam/twitter-mcp — Twitter/X
- @isteam/bluesky-mcp — Bluesky
- @isteam/linkedin-mcp — LinkedIn
License
MIT
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。