MCP AdGuard Home

MCP server for AdGuard Home. Manage DNS filtering, clients, DHCP, rewrites, and more through natural language in Cursor, Claude Code, and Claude Desktop.

Features

• 65 tools across 16 categories covering the complete AdGuard Home API
• Read-only mode via ADGUARD_ACCESS_TIER=read-only for safe monitoring
• Category filtering via ADGUARD_CATEGORIES to expose only the tools you need
• Zero HTTP dependencies -- uses native fetch (Node.js 22+)
• Docker images for linux/amd64 and linux/arm64 on GHCR
• Remote MCP via HTTP transport (MCP_TRANSPORT=http) using the Streamable HTTP protocol
• TypeScript/ESM with full type safety

API Compatibility

Tested with AdGuard Home v0.107.76.

Quick Start

Run the server directly with npx:

ADGUARD_URL="http://your-adguard-ip:3000" \
ADGUARD_USERNAME="your-username" \
ADGUARD_PASSWORD="your-password" \
npx -y @samik081/mcp-adguard-home

The server validates your AdGuard Home connection on startup and fails immediately with a clear error if credentials are missing or invalid.

Docker

Run with Docker (stdio transport, same as npx):

docker run --rm -i \
  -e ADGUARD_URL=http://your-adguard-ip:3000 \
  -e ADGUARD_USERNAME=your-username \
  -e ADGUARD_PASSWORD=your-password \
  ghcr.io/samik081/mcp-adguard-home

To run as a remote MCP server with HTTP transport:

docker run -d -p 3000:3000 \
  -e MCP_TRANSPORT=http \
  -e ADGUARD_URL=http://your-adguard-ip:3000 \
  -e ADGUARD_USERNAME=your-username \
  -e ADGUARD_PASSWORD=your-password \
  ghcr.io/samik081/mcp-adguard-home

The MCP endpoint is available at http://localhost:3000 and a health check at http://localhost:3000/health.

Configuration

Claude Code CLI (recommended):

# Using npx
claude mcp add --transport stdio adguard-home \
  --env ADGUARD_URL=http://your-adguard-ip:3000 \
  --env ADGUARD_USERNAME=your-username \
  --env ADGUARD_PASSWORD=your-password \
  -- npx -y @samik081/mcp-adguard-home

# Using Docker
claude mcp add --transport stdio adguard-home \
  --env ADGUARD_URL=http://your-adguard-ip:3000 \
  --env ADGUARD_USERNAME=your-username \
  --env ADGUARD_PASSWORD=your-password \
  -- docker run --rm -i ghcr.io/samik081/mcp-adguard-home

# Using remote HTTP (connect to a running Docker container or HTTP server)
claude mcp add --transport http adguard-home http://localhost:3000

JSON config (works with Claude Code .mcp.json, Claude Desktop claude_desktop_config.json, Cursor .cursor/mcp.json):

{
  "mcpServers": {
    "adguard-home": {
      "command": "npx",
      "args": ["-y", "@samik081/mcp-adguard-home"],
      "env": {
        "ADGUARD_URL": "http://your-adguard-ip:3000",
        "ADGUARD_USERNAME": "your-username",
        "ADGUARD_PASSWORD": "your-password"
      }
    }
  }
}

Docker (stdio):

{
  "mcpServers": {
    "adguard-home": {
      "command": "docker",
      "args": ["run", "--rm", "-i",
        "-e", "ADGUARD_URL=http://your-adguard-ip:3000",
        "-e", "ADGUARD_USERNAME=your-username",
        "-e", "ADGUARD_PASSWORD=your-password",
        "ghcr.io/samik081/mcp-adguard-home"
      ]
    }
  }
}

Remote MCP (connect to a running Docker container or HTTP server):

{
  "mcpServers": {
    "adguard-home": {
      "type": "streamable-http",
      "url": "http://localhost:3000"
    }
  }
}

Access Tiers

Control which tools are available using the ADGUARD_ACCESS_TIER environment variable:

• full: All 65 tools. Includes configuration changes, adding/removing filters, clearing logs, and all destructive operations.
• read-only: 29 tools. Status, configuration viewing, and query log reading only. No state changes.

Tools that are not available in your tier are not registered with the MCP server. They will not appear in your AI tool's tool list, keeping the context clean.

Environment Variables

Available Categories

global, dns, querylog, stats, filtering, safebrowsing, parental, safesearch, clients, dhcp, rewrites, tls, blocked_services, access, install, mobile_config

Tools

mcp-adguard-home provides 65 tools organized by category. Each tool's Access column shows the minimum tier required: read-only (available in both tiers) or full (requires full tier). The Hints column shows tool behavior: read-only (no state changes), destructive (modifies existing state), idempotent (same result if called twice).