prometheus-mcp
by idanfishman·★ 25·Score 43
MCP server enabling AI agents to query Prometheus monitoring metrics through natural language commands.
Overview
The Prometheus MCP server provides seamless integration between AI assistants and Prometheus monitoring systems. It enables natural language interactions with your monitoring infrastructure by exposing 10 tools across three categories: discovery, info, and query. The server supports both stdio and HTTP transports for maximum compatibility with various MCP clients including VS Code, Cursor, Windsurf, and Claude Desktop.
Try asking AI
After installing, here are 6 things you can ask your AI assistant:
When to choose this
Choose this when you need AI agents to analyze Prometheus monitoring data through natural language queries, especially if you're already using Prometheus and want to enhance it with AI capabilities without additional infrastructure.
When NOT to choose this
Don't choose this if you need write access to your monitoring system or if you're using a monitoring solution other than Prometheus that doesn't expose a compatible API.
Tools this server exposes
10 tools extracted from the READMEprometheus_list_metricsList all available Prometheus metrics
prometheus_metric_metadataGet metadata for a specific Prometheus metric
prometheus_list_labelsList all available Prometheus labels
prometheus_label_valuesGet all values for a specific Prometheus label
prometheus_list_targetsList all Prometheus scrape targets
prometheus_scrape_pool_targetsGet targets for a specific scrape pool
prometheus_runtime_infoGet Prometheus runtime information
prometheus_build_infoGet Prometheus build information
prometheus_queryExecute an instant Prometheus query
prometheus_query_rangeExecute a Prometheus range query
Comparable tools
Installation
Installation
Claude Desktop
- Download the latest
.dxtfile from [releases page](https://github.com/idanfishman/prometheus-mcp/releases) - Double-click to install automatically
- Configure Prometheus URL in extension settings
Or via Developer Settings:
{
"mcpServers": {
"prometheus": {
"command": "npx",
"args": ["prometheus-mcp@latest", "stdio"],
"env": {
"PROMETHEUS_URL": "http://localhost:9090"
}
}
}
}VS Code/Cursor/Windsurf
Add this configuration to your client settings:
{
"mcpServers": {
"prometheus": {
"command": "npx",
"args": ["prometheus-mcp@latest", "stdio"],
"env": {
"PROMETHEUS_URL": "http://localhost:9090"
}
}
}
}FAQ
- What Prometheus server version does this MCP support?
- The Prometheus MCP server works with any Prometheus instance that exposes the Prometheus API (version 1.0 and later).
- Can I use this MCP server with authentication?
- Currently, the server only supports basic Prometheus URL access. For authenticated Prometheus instances, you may need to set up a proxy or tunnel.
- What transport methods are supported?
- The server supports both stdio and HTTP transports, making it compatible with various MCP clients and deployment scenarios.
Compare prometheus-mcp with
Last updated · Auto-generated from public README + GitHub signals.