Datadog MCP Server
Enables integration with Datadog APIs to monitor and retrieve information about monitors, metrics, dashboards, logs, events, and incidents through the Model Context Protocol.
README
Datadog MCP Server
A generic Model Context Protocol (MCP) server for Datadog integration. This project enables Datadog API access via MCP for monitoring, metrics, dashboards, logs, events, and incidents.
Features
- Monitors: List and filter Datadog monitors by status, tags, etc.
- Metrics: Search and retrieve metric information
- Dashboards: List and get dashboard details
- Logs: Search logs with advanced filtering
- Events: Retrieve events within time ranges
- Incidents: List and search incidents
Setup
1. Install Dependencies
npm install
2. Configure Datadog Credentials
- Copy the environment template:
cp .env.template .env
- Get your Datadog API credentials from your Datadog account settings:
- API Key: https://app.datadoghq.com/organization-settings/api-keys
- Application Key: https://app.datadoghq.com/organization-settings/application-keys
- Edit
.envand fill in your credentials:
DD_API_KEY=your_datadog_api_key_here
DD_APP_KEY=your_datadog_app_key_here
DD_SITE=datadoghq.com # or your specific site (datadoghq.eu, us3.datadoghq.com, etc.)
3. Test Configuration
License
This project is licensed under the MIT License. See the LICENSE file for details.
npm test
This will validate your credentials and test the Datadog API connection.
Usage
VS Code MCP Integration
Add to your VS Code MCP configuration (~/.config/Code/User/mcp.json):
{
"mcpServers": {
"datadog": {
"command": "node",
"args": ["/absolute/path/to/datadog/index.js"],
"env": {
"DD_API_KEY": "your_api_key",
"DD_APP_KEY": "your_app_key",
"DD_SITE": "datadoghq.com"
}
}
}
}
Standalone Testing
# List available tools
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/list", "params": {}}' | node index.js
# Test monitors tool
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "get_monitors", "arguments": {"limit": 5}}}' | node index.js
Available Tools
get_monitors
- Description: Retrieve Datadog monitors with filtering options
- Parameters:
limit(optional): Number of monitors to return (default: 10)groupStates(optional): Filter by monitor states (e.g., ["Alert", "Warn"])tags(optional): Filter by tags
get_metrics
- Description: Search for metrics in your Datadog account
- Parameters:
query(optional): Search query for metric nameslimit(optional): Number of metrics to return (default: 100)
get_dashboards
- Description: List Datadog dashboards
- Parameters:
limit(optional): Number of dashboards to return (default: 10)author(optional): Filter by authorfilterDeleted(optional): Filter out deleted dashboards
search_logs
- Description: Search logs with advanced filtering
- Parameters:
query(optional): Log search queryfrom(optional): Start time (e.g., "now-1h")to(optional): End time (e.g., "now")limit(optional): Number of logs to return (default: 100)
get_events
- Description: Retrieve events within a time range
- Parameters:
start(required): Start timestamp (Unix timestamp)end(required): End timestamp (Unix timestamp)priority(optional): Event priority filtertags(optional): Tag filter
get_incidents
- Description: List incidents from Datadog incident management
- Parameters:
query(optional): Search query for incidentslimit(optional): Number of incidents to return (default: 100)
Troubleshooting
Common Issues
-
"spawn datadog-mcp-server ENOENT": This means the official server isn't available. Use this custom server instead.
-
API Authentication Errors:
- Verify your
DD_API_KEYandDD_APP_KEYare correct - Check that your API keys have the necessary permissions
- Ensure you're using the correct
DD_SITEfor your Datadog instance
- Verify your
-
Network/Timeout Errors:
- Check your internet connection
- Verify firewall settings allow HTTPS to Datadog APIs
- Try increasing timeout values if on a slow connection
Debug Mode
Run with debug logging:
DEBUG=1 node index.js
Development
Testing Changes
# Install dependencies
npm install
# Test the server
npm test
# Start in development mode with auto-reload
npm run dev
Adding New Tools
- Define the Zod schema for input validation
- Add the tool to the
listToolshandler - Implement the tool logic in the
callToolhandler - Update this README with documentation
License
MIT License - See LICENSE file for details.
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。
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。
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。
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。
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。
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。
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。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。