AppLovin Max MCP Server

A Model Context Protocol (MCP) server that provides LLM clients like Claude with access to AppLovin Max advertising platform analytics and reporting capabilities.

⚠️ IMPORTANT: PROOF OF CONCEPT ONLY

This is a proof-of-concept throwaway project created for demonstration and experimentation purposes. It is NOT production-ready and should NOT be used in production environments without significant additional work.

Known Limitations:

• No comprehensive error handling or recovery mechanisms
• No rate limiting or request throttling
• No input validation beyond basic parameter checks
• No logging or monitoring capabilities
• No unit tests or integration tests
• No security audits performed
• No performance optimization or scalability considerations
• API keys transmitted in query parameters (consider more secure methods for production)

Use at your own risk. This code is provided as-is with no warranties. See the LICENSE file for details.

Overview

This Go-based MCP server exposes AppLovin Max API functionality through standardized tools, enabling AI assistants to query mobile ad revenue data, analyze user cohorts, and generate insights from your AppLovin monetization metrics.

Features

• Revenue Reporting: Access aggregated mediation statistics including impressions, revenue, eCPM, and fill rates
• Cohort Analysis: Track user lifecycle metrics segmented by installation date
• Flexible Filtering: Filter by application, country, platform, ad type, and more
• Multiple Data Views: Choose between revenue, impression, or session-based cohort analysis
• Time-Series Tracking: Monitor how metrics evolve over time (up to 45 days post-install)

Prerequisites

• Go 1.25.4 or later
• AppLovin Max account with API access
• AppLovin API key

Installation

1. Clone the repository:

git clone <repository-url>
cd Applovin-Max-MCP

2. Install dependencies:

go mod download

3. Build the server:

go build -o bootstrap cmd/main.go

Configuration

Set your AppLovin API key as an environment variable:

export APPLOVIN_API_KEY="your_api_key_here"

Usage

Starting the Server

The MCP server communicates via stdio (standard input/output):

./bootstrap

Integrating with Claude Desktop

Add the following configuration to your Claude Desktop config file:

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

{
  "mcpServers": {
    "applovin-max": {
      "command": "/path/to/Applovin-Max-MCP/bootstrap",
      "env": {
        "APPLOVIN_API_KEY": "your_api_key_here"
      }
    }
  }
}

Available Tools

1. Revenue Report (revenue_report)

Retrieves aggregated mediation statistics from AppLovin Max.

Required Parameters

• start (string): Start date in YYYY-MM-DD format
• end (string): End date in YYYY-MM-DD format
• format (string): Output format - json or csv

Optional Parameters

Columns (array, default: ["day", "application"]):

• Dimensions: day, hour, application, package_name, ad_format, country, platform, network, network_placement, max_placement, max_ad_unit_id, custom_network_name, ad_unit_waterfall_name, device_type, store_id, has_idfa, max_ad_unit_test
• Metrics: impressions, responses, requests, attempts, estimated_revenue, ecpm, fill_rate

Filters:

• filter_application: Filter by application name
• filter_package_name: Filter by package name
• filter_ad_type: Filter by ad type (e.g., banner, inter, rewarded)
• filter_country: Filter by two-letter ISO country code
• filter_platform: Filter by platform (android, ios)
• filter_network: Filter by network name
• filter_zone: Filter by zone

Sorting:

• sort_day: Sort by day (ASC/DESC)
• sort_hour: Sort by hour (ASC/DESC)
• `sort_estimat

... View full README on GitHub