OpenTweet MCP Server
Provides 30 tools to post, schedule, thread, and analyze X (Twitter) content, including AI media generation and multi-account management. No X developer account or OAuth setup required.
README
OpenTweet MCP Server
Post, schedule, and thread on X (Twitter) from Claude, Cursor, ChatGPT, Windsurf, or any MCP client. 30 tools covering tweets, threads, scheduling, an evergreen queue, analytics, AI media generation, and multi-account management.
No X developer account, no OAuth app setup, no pay-per-call API billing. OpenTweet owns the X connection; your agent gets simple tools.
<p align="center"> <img src="icon.png" alt="OpenTweet" width="128" height="128"> </p>
Quick start: hosted server (recommended)
Nothing to install. Point your MCP client at the hosted Streamable HTTP endpoint:
https://mcp.opentweet.io/mcp
Claude Code, one command:
claude mcp add --transport http opentweet \
https://mcp.opentweet.io/mcp \
--header "Authorization: Bearer ot_your_api_key"
Or any MCP client config (Claude, Cursor, Windsurf, ChatGPT connectors):
{
"mcpServers": {
"opentweet": {
"type": "streamable-http",
"url": "https://mcp.opentweet.io/mcp",
"headers": { "Authorization": "Bearer ot_your_api_key" }
}
}
}
Get your API key (prefix ot_) at opentweet.io/developer. Connect your X account once in the dashboard; the key covers everything else.
Other install options
Claude Desktop extension (one click)
Download the .mcpb extension from opentweet.io/claude-desktop and double-click it. Your API key is stored in the OS keychain.
Local via npm
npx -y @opentweet/mcp-server install # interactive installer, detects your client
Or add manually:
{
"mcpServers": {
"opentweet": {
"command": "npx",
"args": ["-y", "@opentweet/mcp-server"],
"env": { "OPENTWEET_API_KEY": "ot_your_api_key" }
}
}
}
Config file locations:
- Claude Desktop:
~/Library/Application Support/Claude/claude_desktop_config.json(macOS) or%APPDATA%\Claude\claude_desktop_config.json(Windows) - Cursor:
~/.cursor/mcp.json - Windsurf:
~/.codeium/windsurf/mcp_config.json
The 30 tools
Tweets and threads
create_tweet, create_thread, list_tweets, get_tweet, update_tweet, delete_tweet, publish_tweet, schedule_tweet, batch_schedule (up to 50 at once), repurpose_tweet
Media and AI generation
upload_media, generate_image, generate_video
Analytics and insights
get_analytics, get_top_posts, get_profile_growth, get_best_times, analyze_best_times, get_content_gaps, search_inspiration
Account
get_account, list_accounts (multi-account: pass x_account_id to any tool)
Evergreen queue
get_evergreen_settings, update_evergreen_settings, list_evergreen_posts, get_evergreen_post, add_to_evergreen, update_evergreen_post, remove_from_evergreen, get_evergreen_history
Plus MCP resources (account status, scheduled queue) and built-in prompts.
Example prompts
- "Write 3 tweets about TypeScript tips and schedule them across this week at my best posting times."
- "Turn this blog post into a thread and schedule it for tomorrow 9am."
- "What were my top 5 tweets this month? Add the best one to my evergreen queue."
- "Check my posting streak and content gaps, then draft tweets to fill them."
How it compares
X's official MCP server (XMCP) exposes the raw X API and can post, but it requires an X developer account, OAuth app credentials, and pay-per-use API billing, and it has no scheduling. OpenTweet is the managed publishing layer: flat pricing from $11.99/month, scheduling, threads, evergreen recycling, and analytics included. Full comparison: opentweet.io/mcp/post-to-x.
Links
- Setup guide: opentweet.io/twitter-mcp-server
- REST API docs: opentweet.io/docs/api
- Pricing: opentweet.io/pricing (7-day free trial)
- npm: @opentweet/mcp-server
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。