talkopsai/alertmanager-mcp-server repository overview

⁠Alertmanager MCP Server

A comprehensive Model Context Protocol (MCP) server that gives AI assistants the power to triage alerts, manage silences, inspect routing, and govern Alertmanager operations — from on-call summarization to notification pipeline testing.

⁠✨ Features

• On-Call Triage: AI summarizes active alerts grouped by severity and service, explains routing paths, and identifies alerts falling into the default route — all in one guided workflow.
• Safe Silence Management: Mandatory preview dry-runs before creating silences, duplicate detection, 24-hour duration caps, blast-radius warnings, and policy validation — preventing overly broad silences that could mask real incidents.
• Routing Introspection: Simulate routing for any label set, inspect the full routing tree, list receivers with integration type detection (Slack, PagerDuty, email, webhook), and audit default-route alerts.
• Governance & Compliance: Export effective configuration for Git storage, audit recent silence changes with author tracking, and validate proposed silences against organizational policy.
• Multi-Backend Support: Manage multiple Alertmanager backends with explicit backend_id on every call — no hidden defaults.
• Integration Testing: Push test alerts to verify notification pipelines end-to-end.

⁠🚀 How to Use This Image

This Docker image runs an MCP server that connects to your Alertmanager instance(s). AI assistants (Claude, Cline, Cursor, or any MCP client) connect to the server over HTTP, SSE, or stdio.

⁠Running Standalone (HTTP Transport)

docker run --rm -it \
  -p 8769:8769 \
  -e ALERTMANAGER_BASE_URL=http://host.docker.internal:9093 \
  -e MCP_TRANSPORT=http \
  talkopsai/alertmanager-mcp-server:latest

The server is now listening on http://localhost:8769/mcp.

Then configure your MCP client to connect:

{
  "mcpServers": {
    "alertmanager": {
      "url": "http://localhost:8769/mcp",
      "description": "MCP Server for Alertmanager alert triage, silence management, and routing"
    }
  }
}

⁠Multi-Backend Setup

Connect to multiple Alertmanager backends simultaneously:

docker run --rm -it \
  -p 8769:8769 \
  -e MCP_TRANSPORT=http \
  -e ALERTMANAGER_BACKENDS='[
    {"id": "prod", "base_url": "https://alertmanager-prod.example.com"},
    {"id": "staging", "base_url": "https://alertmanager-staging.example.com"}
  ]' \
  talkopsai/alertmanager-mcp-server:latest

⁠🛠️ Configuration (Claude Desktop — Stdio)

To use this image over stdio transport directly inside Claude Desktop:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "alertmanager-mcp-server": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e", "ALERTMANAGER_BASE_URL=http://host.docker.internal:9093",
        "-e", "MCP_TRANSPORT=stdio",
        "-e", "MCP_LOG_LEVEL=INFO",
        "talkopsai/alertmanager-mcp-server:latest"
      ]
    }
  }
}

⁠⚙️ Environment Variables

⁠Server Configuration

⁠Alertmanager Backend

⁠Silence Safety

⁠🏷️ Supported Tags

• latest — Latest stable release
• 0.1.0 — Initial release

⁠🔧 Available Tools (14 Total)

⁠Alert Triage

⁠Silence Lifecycle

⁠Silence Helpers

⁠Routing & Notifications

⁠Governance & Audit

⁠On-Call Triage

⁠📖 Available Resources (11 URIs)

⁠🔐 Security Best Practices

1. Never expose the MCP server to the public internet without proper authentication.
2. Silences affect real alert notifications — always use am_preview_silence before creating silences in production.
3. Test alerts fire real notifications — am_push_test_alert will trigger downstream integrations (Slack, PagerDuty, email).
4. Configuration export may contain sensitive routing rules — treat exported configs as confidential.
5. Silence duration caps — the default 24h cap prevents forgotten silences from masking incidents.

⁠🐳 Docker Compose Example

services:
  alertmanager-mcp-server:
    image: talkopsai/alertmanager-mcp-server:latest
    ports:
      - "8769:8769"
    environment:
      - ALERTMANAGER_BASE_URL=http://alertmanager:9093
      - MCP_TRANSPORT=http
      - MCP_LOG_LEVEL=INFO
    depends_on:
      - alertmanager

  alertmanager:
    image: prom/alertmanager:latest
    ports:
      - "9093:9093"
    volumes:
      - ./alertmanager.yml:/etc/alertmanager/alertmanager.yml

⁠🔗 Links and Resources

• Source Code: GitHub Repository⁠
• Full Documentation: README & Guides⁠
• Prometheus MCP Server: Companion for metrics & PromQL⁠
• Model Context Protocol: MCP Official Site⁠
• Alertmanager: Alertmanager Documentation⁠
• Report Issues: GitHub Issues⁠
• Discord: Join the Community⁠
• License: Apache 2.0⁠

If you find this MCP server useful, consider leaving a ⭐ on the GitHub repository⁠!