MCP Server (AI Integration)
Seamlessly integrate your telemetry data with advanced Large Language Models like Claude or Cursor IDE. The Model Context Protocol (MCP) server exposes 50+ read-only tools spanning every pillar — APM, RUM, Logs, Uptime, Infrastructure, Databases, Queues, Errors, Alerts & Incidents, AI Monitoring and Billing — plus built-in investigation prompts (root-cause, incident triage, AI cost review) and addressable resources, so an agent can query, correlate and summarize your observability data in natural language. Connections use the modern Streamable HTTP transport.
Prerequisites
- An active Senzor workspace with existing telemetry.
- An MCP-compatible client (Cursor IDE, Claude Desktop, or any Streamable HTTP client).
1. Dashboard Configuration
Before injecting code into your application, you must provision an API key from the Senzor dashboard.
Generate MCP Key
Navigate to AI Integrations in the dashboard and generate a dedicated MCP API Key. Keys are hashed at rest, shown once, and revocable at any time.
Connect & Explore
Add the server to your client using the configuration below. Your agent can then list every tool, read resources, and run the built-in investigation prompts as slash-commands.
2. SDK Installation
Select your environment below to view the initialization code.
{
"mcpServers": {
"senzor": {
"url": "https://api.senzor.dev/api/mcp",
"headers": {
"Authorization": "Bearer <YOUR_MCP_KEY>"
}
}
}
}Add this configuration to your Cursor settings under Features > MCP. Cursor auto-detects the Streamable HTTP transport.