mcpField
P
pmcp

pmcp

Local
github.com/yshnggData & Analytics4Updated 14w ago
Repository

About pmcp

pmcp is a Go-based MCP server that exposes Prometheus query and management capabilities. It is for MCP clients that need to query metrics, inspect metadata, or access server operations and data.

Key features

  1. 01Instant and range PromQL queries
  2. 02Metric and label metadata lookup
  3. 03Target and alert discovery
  4. 04TSDB administration tools
  5. 05Prometheus status and runtime resources

Use cases

  1. 01Querying metrics from an MCP client
  2. 02Inspecting targets, alerts, and rule metadata
  3. 03Checking Prometheus runtime and TSDB details

MCP support

The server exposes Prometheus tools, resources, and a metric-listing prompt to MCP clients. It supports stdio and HTTP transports, with optional bearer-token authentication for HTTP.

#pmcp#prometheus#monitoring

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "pmcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "pull"
      ]
    }
  }
}
Run commanddocker run -i --rm pull

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
ocipull

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Prometheus MCP Server

codecov

A Go-based MCP server that exposes Prometheus query capabilities via the Model Context Protocol.

Installation

go install github.com/yshngg/prometheus-mcp-server@latest

Or pull the Docker image:

docker pull ghcr.io/yshngg/prometheus-mcp-server:latest

Usage

prometheus-mcp-server --prom-addr="http://localhost:9090"
Flags
FlagDefaultDescription
-prom-addrhttp://localhost:9090Prometheus server URL
-mcp-addrlocalhost:8080MCP server listen address
-transportstdiostdio or http
-auth-token``Bearer token for MCP endpoint authentication (optional)
-versionPrint version
Environment Variables

All flags can be set via environment variables: PROM_ADDR, MCP_ADDR, TRANSPORT, AUTH_TOKEN.

MCP Client Configuration

Add the following config to your MCP client:

{
  "mcpServers": {
    "prometheus": {
      "command": "prometheus-mcp-server",
      "args": ["--prom-addr", "http://localhost:9090"]
    }
  }
}

Note

Install the binary via go install github.com/yshngg/prometheus-mcp-server@latest or pull the Docker image with docker pull ghcr.io/yshngg/prometheus-mcp-server:latest.

If you don't have the binary installed, use Docker:

{
  "mcpServers": {
    "prometheus": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "ghcr.io/yshngg/prometheus-mcp-server:latest",
        "--prom-addr",
        "http://host.docker.internal:9090"
      ]
    }
  }
}
Client-specific configuration
Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "prometheus": {
      "command": "prometheus-mcp-server",
      "args": ["--prom-addr", "http://localhost:9090"]
    }
  }
}

With Docker:

{
  "mcpServers": {
    "prometheus": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "ghcr.io/yshngg/prometheus-mcp-server:latest",
        "--prom-addr",
        "http://host.docker.internal:9090"
      ]
    }
  }
}
OpenCode

Add to opencode.json (guide):

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "prometheus": {
      "type": "local",
      "command": [
        "prometheus-mcp-server",
        "--prom-addr",
        "http://localhost:9090"
      ]
    }
  }
}
HTTP Transport

Run the server with HTTP transport:

prometheus-mcp-server --transport=http --mcp-addr="localhost:8080"

Then configure your client to use http://localhost:8080/mcp.

For authenticated access, set AUTH_TOKEN:

AUTH_TOKEN=my-token prometheus-mcp-server --transport=http

Your client must then include Authorization: Bearer my-token in requests to /mcp.

Tools

Expression: instant_query, range_query

Metadata: find_series_by_labels, list_label_names, list_label_values, target_metadata_query, metric_metadata_query

Discovery: target_discovery, alert_query, rule_query, alertmanager_discovery

TSDB Admin: tsdb_snapshot, delete_series, clean_tombstones

Management: health_check, readiness_check, reload, quit

Resources

The server exposes Prometheus data as URI-addressable resources under the prom:/// scheme:

URIDescription
prom:///configCurrently loaded configuration (YAML)
prom:///flagsCommand-line flag values
prom:///runtime-infoRuntime information
prom:///build-infoBuild information
prom:///tsdb-statsTSDB cardinality statistics
prom:///wal-replay-statsWAL replay state
Resource Templates

URI templates for dynamic data access:

TemplateDescription
prom:///api/v1/query?query={promql}Instant PromQL query result
prom:///api/v1/label/{name}/valuesLabel values for a given label name

Prompts

NameArgumentsDescription
all-available-metricsprefix (optional)Lists all metric names in the Prometheus instance

License

Apache License 2.0

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.