What Prometheus operations does this MCP server support?

The server provides API-complete access to Prometheus-compatible backends, enabling metric queries, alert management, and rule creation. Consult the server documentation for the full list of supported operations and endpoints.

How do I authenticate with my Prometheus instance?

Authentication details depend on your Prometheus setup. Check the environment variables required by the Docker image (ghcr.io/tjhop/prometheus-mcp-server:0.18.0) and your Prometheus backend's security configuration.

Can I use this with any Prometheus-compatible backend?

Yes, the server is designed to work with Prometheus-compatible backends. You'll need to configure the appropriate connection details via environment variables when running the Docker container.

Is there a cost to using this MCP server?

The server itself is open-source and free. Costs depend on your Prometheus infrastructure and any cloud monitoring services you use.

What's the difference between this and querying Prometheus directly?

This MCP server integrates Prometheus as a managed resource within AI chat workflows, allowing Claude or other AI models to query metrics and manage alerts without direct API calls. It abstracts authentication and provides a standardized interface.

MCP ServerSTDIOOfficialv0.18.0

Prometheus MCP Server

Query and manage Prometheus-compatible monitoring backends via MCP. For developers integrating observability APIs into AI workflows or building monitoring automation tools.

io.github.tjhop/prometheus-mcp-server

Try MCP with AI models View source

Hosted URL

Local install

Transport

STDIO

Auth

No auth required

What the Prometheus MCP server does

How models use it and what it is built for.

Query and manage Prometheus-compatible monitoring backends via MCP. For developers integrating observability APIs into AI workflows or building monitoring automation tools.

Connect to Prometheus

Local install — runs as a subprocess.

docker run ghcr.io/tjhop/prometheus-mcp-server:0.18.0

Resources

Where to find authoritative docs and source for Prometheus.

Source repository

Example prompts for Prometheus

Paste any of these into Agent Studio after connecting Prometheus.

Query CPU usage metrics from the last hour across all instances
List all active alerts and their current status in Prometheus
Create a new recording rule that aggregates request latency percentiles
What environment variables does this server need to connect to my Prometheus instance?

Prometheus MCP server — FAQ

Common questions about connecting and running Prometheus.

What Prometheus operations does this MCP server support?
The server provides API-complete access to Prometheus-compatible backends, enabling metric queries, alert management, and rule creation. Consult the server documentation for the full list of supported operations and endpoints.
How do I authenticate with my Prometheus instance?
Authentication details depend on your Prometheus setup. Check the environment variables required by the Docker image (ghcr.io/tjhop/prometheus-mcp-server:0.18.0) and your Prometheus backend's security configuration.
Can I use this with any Prometheus-compatible backend?
Yes, the server is designed to work with Prometheus-compatible backends. You'll need to configure the appropriate connection details via environment variables when running the Docker container.
Is there a cost to using this MCP server?
The server itself is open-source and free. Costs depend on your Prometheus infrastructure and any cloud monitoring services you use.
What's the difference between this and querying Prometheus directly?
This MCP server integrates Prometheus as a managed resource within AI chat workflows, allowing Claude or other AI models to query metrics and manage alerts without direct API calls. It abstracts authentication and provides a standardized interface.

Skip the local setup — run MCP in your browser

MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 60+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.

Open Agent Studio