recommended-by-mcp
Stdio MCP server for managing recommended.by lists from agents like OpenClaw, Codex, Claude Desktop, etc. It exposes tools to create, read, update, delete lists and items via the recommended.by REST API.
README
recommended-by-mcp
Stdio MCP server for managing your recommended.by lists from agents like OpenClaw, Codex, Claude Desktop, Claude Code, Cursor, and Windsurf.
The package talks to the recommended.by REST API and exposes MCP tools over stdio. This is useful for clients that do not support remote HTTP MCP headers consistently: the agent launches this package locally, and the package handles recommended.by authentication.
Quick Start
Create an API key in Dashboard -> Profile & URL -> API Keys, then authenticate the package once:
npx recommended-by-mcp@latest login rb_live_your_key_here
Then configure your MCP client to run:
npx -y recommended-by-mcp@latest
You can also avoid local storage and pass the key through an environment variable:
RECOMMENDED_BY_API_KEY=rb_live_your_key_here npx -y recommended-by-mcp@latest
Authentication
The server resolves auth in this order:
--api-key rb_live_...RECOMMENDED_BY_API_KEY- saved config from
recommended-by-mcp login
The saved config lives at:
- macOS:
~/Library/Application Support/recommended-by-mcp/config.json - Linux:
~/.config/recommended-by-mcp/config.json - Windows:
%APPDATA%\recommended-by-mcp\config.json
Remove the saved key with:
npx recommended-by-mcp@latest logout
Check which account the key can access:
npx recommended-by-mcp@latest whoami
MCP Tools
recommended_list_listsrecommended_create_listrecommended_get_listrecommended_update_listrecommended_delete_listrecommended_list_itemsrecommended_search_itemsrecommended_add_item_from_resultrecommended_add_itemrecommended_update_itemrecommended_delete_item
Use recommended_search_items before adding ambiguous books, places, podcasts, music, movies, or TV. It returns candidates with stable sourceId values plus preview fields. For Google Places candidates, show confirmationUrl and previewImageUrl to the user before calling recommended_add_item_from_result.
Client Examples
Codex
Add this to ~/.codex/config.toml:
[mcp_servers.recommended_by]
command = "npx"
args = ["-y", "recommended-by-mcp@latest"]
env = { RECOMMENDED_BY_API_KEY = "rb_live_your_key_here" }
OpenClaw
openclaw mcp add recommended-by \
--command npx \
--arg -y \
--arg recommended-by-mcp@latest \
--env RECOMMENDED_BY_API_KEY=rb_live_your_key_here
openclaw mcp doctor recommended-by --probe
Claude Desktop
Add this to claude_desktop_config.json:
{
"mcpServers": {
"recommended-by": {
"command": "npx",
"args": ["-y", "recommended-by-mcp@latest"],
"env": {
"RECOMMENDED_BY_API_KEY": "rb_live_your_key_here"
}
}
}
}
Development
pnpm install
pnpm --filter recommended-by-mcp build
pnpm --filter recommended-by-mcp test
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。