Sagu MCP Server
Connects AI agents to Sagu radars, pings, and artifacts for intelligence operations, enabling radar search, context packet export, and artifact management.
README
<p align="center"> <img src="assets/logo.png" alt="Sagu" width="88" height="88" /> </p>
<h1 align="center">Sagu plugins</h1>
<p align="center"><strong>Agent-facing intelligence from your radars.</strong></p>
<p align="center"> Connect agents to <a href="https://sagu.app">Sagu</a> radars, pings, and artifacts over MCP. List and search radars, export agent-ready context packets, refresh intelligence on demand, and create or fetch platform artifacts. </p>
<p align="center"> <a href="https://sagu.app">sagu.app</a> · MCP: <code>https://sagu-mcp.luke-nittmann.workers.dev/mcp</code> </p>
Install
<!-- AUTO-GENERATED:INSTALL START -->
Any agent (npx skills)
Works across Claude Code, Cursor, Codex, Copilot, Windsurf, and 10+ skill-aware agents.
npx skills add creative-int/sagu-plugins
Claude Code
Add the marketplace, then install the Sagu plugin.
/plugin marketplace add creative-int/sagu-plugins
/plugin install sagu@sagu
Codex
Add this repo as a Codex plugin marketplace, then install from /plugins.
codex plugin marketplace add creative-int/sagu-plugins
Cursor
Install Sagu from the Cursor plugin marketplace.
Cursor → Settings → Plugins → Add marketplace → creative-int/sagu-plugins
Any remote MCP client (Cline, Windsurf, Zed, VS Code, generic)
Point your client at the remote Sagu MCP server.
{
"mcpServers": {
"sagu": {
"url": "https://sagu-mcp.luke-nittmann.workers.dev/mcp"
}
}
}
Local stdio MCP client
Run the Sagu stdio MCP server locally with your Sagu API key.
{
"mcpServers": {
"sagu": {
"command": "npx",
"args": [
"-y",
"@sagu/mcp"
],
"env": {
"RADAR_API_KEY": "sagu_sk_your_key_here"
}
}
}
}
<!-- AUTO-GENERATED:INSTALL END -->
To preview the available skills without installing:
npx skills add creative-int/sagu-plugins --list
Included skills
sagu-dispatch— list and search radars, read pings, export agent-ready context packets, refresh intelligence, and ask questions in radar threads. Also answers tomonitorandradar.sagu-artifact— create and fetch platform artifacts, extract format-matched dispatch artifacts from pings, and evolve radar knowledge. Also answers toartifactandshare.
Skills live under skills/.
MCP tools
See the tool surface documented in sagu.config.ts. Tools are
hosted by the Sagu MCP server (packages/mcp stdio + apps/mcp remote) and
include radars, pings, artifacts, research, and knowledge-graph operations.
Develop
pnpm install
pnpm generate # regenerate all adapters from sagu.config.ts
pnpm verify # drift check + typecheck + MCP reachability smoke
License
MIT © creative-int
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。