redfox-mcp
Provides 40 MCP tools for searching and retrieving content from six major platforms (Douyin, Xiaohongshu, WeChat, Bilibili, Toutiao, TikTok) along with AI search and image/video generation via RedFoxHub's API.
README
<div align="center"> <a href="https://pypi.org/project/redfox-python-sdk/"><img src="https://img.shields.io/pypi/v/redfox-mcp.svg" alt="PyPI version"></a> <a href="https://pypi.org/project/redfox-mcp/"><img src="https://img.shields.io/pypi/pyversions/redfox-mcp.svg" alt="Python"></a> <a href="https://pypi.org/project/redfox-mcp/"><img src="https://img.shields.io/pypi/l/redfox-mcp.svg" alt="License"></a>
<p align="center"> <a href="https://github.com/redfox-data/redfox-mcp/blob/main/README.zh.md">中文</a> English </p>
</div>
<p align="center"> <a href="https://redfox.hk/?source=mcp"><img src="https://lyy.redfox.hk/page/logo-redfox-name.png" alt="RedFox Logo" width="200"></a> </p>
redfox-mcp
RedFoxHub MCP Server — turns the data APIs of 6 major content platforms plus AI search / generation capabilities into 40 MCP tools, ready for any MCP client such as dsh, Claude Code or Cursor.
Tools (40)
| Category | Tools | Notes |
|---|---|---|
| Douyin | douyin_search_articles / douyin_search_users / douyin_get_user / douyin_get_user_works / douyin_get_work / douyin_search_ai_articles |
work search, account search, account info, work lists, work detail, AI-work feed |
| Xiaohongshu | xiaohongshu_search_articles / xiaohongshu_search_users / xiaohongshu_get_account / xiaohongshu_get_work / xiaohongshu_search_ai_articles |
note search, creator search, account info, note detail, AI-note feed |
| WeChat Official Accounts | wechat_search_articles / wechat_search_users / wechat_get_account / wechat_get_user_works / wechat_get_work / wechat_get_article_detail / wechat_search_ai_articles |
article search (incl. full-text detail), account search, article lists, AI-article feed |
| Bilibili | bilibili_search_articles / bilibili_search_users / bilibili_get_account / bilibili_get_user_works / bilibili_get_work |
video search, UP-master search / info / videos, video detail |
| Toutiao | toutiao_search_works / toutiao_get_work |
content search, work detail (realtime) |
| TikTok | tiktok_search_users |
account search |
| AI search | ai_search_kimi / ai_search_doubao / ai_search_deepseek |
one call submits the query and waits for the full answer |
| AI generation | gpt_image_generate / doubao_image_pro_generate / doubao_image_lite_generate / doubao_video_generate |
text-to-image / image-to-image / image sets / text-to-video, one call submits and waits |
| Task follow-up | ai_search_*_result / gpt_image_result / doubao_image_*_result / doubao_video_result (7 in total) |
when an async tool times out it returns a taskId; use the matching result tool to fetch the outcome |
Async tools (AI search / generation) poll internally: submit → wait → return the full result, with no manual taskId handling. If the wait exceeds timeout_seconds (default 240s, 480s for video), a taskId is returned for the matching result tool.
Authentication
All APIs require a RedFoxHub API key:
- Get one at https://redfox.hk/settings/api-keys/?source=mcp
- Set the environment variable:
export REDFOX_API_KEY="YOUR_API_KEY"
Without a key, every tool returns a structured message explaining how to obtain one.
Install & Run
Python ≥ 3.10 required. Recommended via uv:
uvx redfox-mcp
or:
pip install redfox-mcp
redfox-mcp
The server runs on stdio transport.
Client Configuration
dsh (DeepSeek Harness)
Install the official bundle plugin redfox-community-dsh — it registers this MCP server out of the box, exposing tools as mcp__redfox__*:
dsh plugin --profile web add -w github:redfox-data/redfox-community-dsh
Claude Code
claude mcp add redfox --env REDFOX_API_KEY=YOUR_API_KEY -- uvx redfox-mcp
Cursor / other MCP clients
Add to your MCP configuration:
{
"mcpServers": {
"redfox": {
"command": "uvx",
"args": ["redfox-mcp"],
"env": { "REDFOX_API_KEY": "YOUR_API_KEY" }
}
}
}
Under the Hood
Built on the official SDK redfox-python-sdk. API docs: https://redfox.hk/?source=mcp.
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。