Scout MCP Server

Scout MCP Server

Enables AI agents to interact with Scout Live platform capabilities through standardized MCP primitives, including tools for app management, deployment, and logging.

Category
访问服务器

README

Scout MCP Server

An MCP (Model Context Protocol) server that wraps the Scout SDK, enabling AI agents to interact with Scout Live platform capabilities through standardized MCP primitives.

What is MCP?

The Model Context Protocol is an open protocol that enables seamless integration between LLM applications and external data sources and tools. It provides:

  • Resources: Context and data for AI models
  • Prompts: Templated messages and workflows
  • Tools: Functions for AI models to execute

Features

This MCP server exposes Scout Live capabilities as:

Tools (Actions)

Tool Description Auth Required
app_list List user's apps User
app_create Create a new app User
deploy Deploy app version App (owner)
logs Get app logs App
port_get Get port adapter config App
port_set Set port adapter mapping App (owner)

Resources (Data)

Resource Description Auth Required
scout://apps List of all apps User
scout://adapters Available adapters Public
scout://app/{slug} App details App

Prompts (Workflows)

Prompt Description
deploy-new-app Create and deploy a new app
debug-errors Investigate app errors

Installation

# Clone the repository
git clone https://github.com/scoutos-labs/scout-mcp-server.git
cd scout-mcp-server

# Install dependencies
bun install

# Build
bun run build

# Run with stdio transport (local use)
bun run start

# Run with HTTP transport (server mode)
bun run start:http

Configuration

Environment Variables

Copy .env.example to .env and configure:

Variable Description Required
SCOUT_API_KEY Scout Live API key Yes
SCOUT_API_URL API endpoint (default: https://scoutos.live) No
MCP_TRANSPORT Transport mode: stdio or http No (default: stdio)
PORT HTTP server port No (default: 3000)

For Claude Desktop

Add to your Claude Desktop config:

{
  "mcpServers": {
    "scout": {
      "command": "bun",
      "args": ["run", "/path/to/scout-mcp-server/dist/index.js"],
      "env": {
        "SCOUT_API_KEY": "your-api-key"
      }
    }
  }
}

For Claude Code

{
  "mcpServers": {
    "scout": {
      "command": "bun",
      "args": ["run", "/path/to/scout-mcp-server/dist/index.js"],
      "env": {
        "SCOUT_API_KEY": "your-api-key"
      }
    }
  }
}

Development

# Run tests
bun run test

# Run tests in watch mode
bun run test:watch

# Run with coverage
bun run test:coverage

# Lint
bun run lint

Deployment to scoutos.live

Prerequisites

  1. DigitalOcean Container Registry access
  2. kubectl configured for scout-live cluster
  3. Scout Live namespace exists

Deploy

# Build and deploy
./deploy.sh v0.1.0

# Or manual steps:

# 1. Build
bun run build

# 2. Build Docker image
docker build -t registry.digitalocean.com/scout-live/mcp-server:v0.1.0 .

# 3. Push to registry
docker push registry.digitalocean.com/scout-live/mcp-server:v0.1.0

# 4. Apply Kubernetes manifests
kubectl apply -f k8s/deployment.yaml -n scout-live
kubectl apply -f k8s/ingress.yaml -n scout-live

# 5. Update deployment image
kubectl set image deployment/scout-mcp-server \
  mcp-server=registry.digitalocean.com/scout-live/mcp-server:v0.1.0 \
  -n scout-live

Verify Deployment

# Check pod status
kubectl get pods -n scout-live -l app=scout-mcp-server

# Check logs
kubectl logs -f deployment/scout-mcp-server -n scout-live

# Health check
curl https://mcp.scoutos.live/health

Access Endpoints

After deployment, the MCP server is available at:

Endpoint Description
https://mcp.scoutos.live/ Server capabilities
https://mcp.scoutos.live/health Health check
https://mcp.scoutos.live/sse SSE transport endpoint

Architecture

┌──────────────────────────────────────────────┐
│               MCP Host (Claude, etc.)          │
└─────────────────────────┬────────────────────┘
                          │
                          ▼
┌──────────────────────────────────────────────┐
│           Scout MCP Server (this repo)        │
│                                               │
│  ┌─────────┐  ┌─────────┐  ┌─────────────┐   │
│  │ Tools   │  │Resources│  │ Prompts    │   │
│  └────┬────┘  └────┬────┘  └──────┬──────┘   │
│       │            │              │          │
│       └────────────┼──────────────┘          │
│                    │                         │
└────────────────────┼─────────────────────────┘
                     │
                     ▼
┌──────────────────────────────────────────────┐
│              Scout Live API                   │
│              (scoutos.live)                   │
└──────────────────────────────────────────────┘

API Reference

Tools

app_list

List all apps for the authenticated user.

{
  "name": "app_list",
  "arguments": {}
}

app_create

Create a new app on Scout Live.

{
  "name": "app_create",
  "arguments": {
    "name": "My App",
    "slug": "my-app" // optional, auto-generated if not provided
  }
}

deploy

Deploy a new version of an app.

{
  "name": "deploy",
  "arguments": {
    "slug": "my-app",
    "files": {
      "index.html": "<html>...</html>",
      "style.css": "body { ... }"
    }
  }
}

logs

Get recent logs for an app.

{
  "name": "logs",
  "arguments": {
    "slug": "my-app",
    "lines": 100 // optional
  }
}

Resources

scout://apps

List all apps.

scout://adapters

List available port adapters.

scout://app/{slug}

Get details for a specific app.

Prompts

deploy-new-app

Create and deploy a new app.

{
  "name": "deploy-new-app",
  "arguments": {
    "name": "My App",
    "type": "static" // optional: static, hono, express
  }
}

debug-errors

Investigate errors in app logs.

{
  "name": "debug-errors",
  "arguments": {
    "slug": "my-app"
  }
}

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选