
RevenueCat to Adapty Migration MCP
A Model Context Protocol server that helps users migrate subscription businesses from RevenueCat to Adapty through natural language interactions with LLMs like Claude Desktop.
README
RevenueCat to Adapty Migration MCP
A Model Context Protocol (MCP) server that helps users migrate their subscription business from RevenueCat to Adapty. This MCP provides tools for data export, analysis, and automated migration through natural language interactions with LLMs like Claude Desktop.
Features
- RevenueCat Data Export: Hierarchical export of projects, offerings, packages, products, and entitlements
- Migration Analysis: Assess migration complexity and generate recommendations
- Automated Migration: Create Adapty apps, access levels, and products from RevenueCat data
- Validation: Verify migration success and data integrity
- LLM Integration: Natural language interface for migration tasks
Installation
Prerequisites
- Python 3.8 or higher
- RevenueCat API key (v2)
- Adapty account (email/password for authentication)
Local Installation
-
Clone the repository:
git clone <repository-url> cd rc-adapty-migration-mcp
-
Create a virtual environment:
python3 -m venv .venv source .venv/bin/activate # On Windows: .venv\Scripts\activate
-
Install the package:
pip install -e .
-
Set up environment variables:
export REVENUECAT_API_KEY="your_revenuecat_api_key"
Configuration
Environment Variables
REVENUECAT_API_KEY
: Your RevenueCat secret API key (required)REVENUECAT_PROJECT_ID
: Optional RevenueCat project IDLOG_LEVEL
: Optional logging level (INFO, DEBUG, etc.)
Adapty Authentication
Important: Adapty uses username/password authentication, not API keys. Your Adapty credentials are provided at runtime when calling migration tools and are not stored permanently.
MCP Configuration
Create an MCP configuration file (e.g., mcp-config.json
):
{
"mcpServers": {
"rc-adapty-migration": {
"command": "rc-adapty-migration-mcp",
"env": {
"REVENUECAT_API_KEY": "your_revenuecat_api_key"
}
}
}
}
Usage
With Claude Desktop
- Configure Claude Desktop to use this MCP server
- Ask natural language questions like:
- "Export my RevenueCat data and analyze the structure"
- "Migrate my RevenueCat setup to Adapty using my email and password"
- "Validate that my migration was successful"
Available Tools
1. Export RevenueCat Data
{
"name": "export_revenuecat_data",
"arguments": {
"output_format": "summary"
}
}
2. Analyze RevenueCat Structure
{
"name": "analyze_revenuecat_structure",
"arguments": {
"analysis_depth": "detailed"
}
}
3. Migrate to Adapty
{
"name": "migrate_to_adapty",
"arguments": {
"adapty_email": "user@example.com",
"adapty_password": "your_password",
"migration_scope": "full"
}
}
Security Note: When using the migration tool, your Adapty credentials are only used for the specific migration operation and are not stored or logged.
Deployment Options
Option 1: Local Installation (Recommended for Testing)
Users install the MCP locally on their machine:
- Install the package using pip
- Configure environment variables (RevenueCat API key only)
- Point their LLM to the local installation
- Provide Adapty credentials at runtime when needed
Pros:
- Full control over data and credentials
- No network dependencies
- Easy to debug and customize
- No hosting costs
- Adapty credentials never stored
Cons:
- Requires technical setup
- Users must manage their own environment
Option 2: Hosted Service
Deploy the MCP as a hosted service:
- Deploy to cloud platform (AWS, GCP, Azure)
- Set up authentication for users
- Provide web interface or API endpoints
Pros:
- No local setup required
- Centralized management
- Easy updates and monitoring
Cons:
- Requires hosting infrastructure
- Security considerations for user credentials
- Network dependencies
Development
Project Structure
rc-adapty-migration-mcp/
├── src/rc_adapty_migration/
│ ├── api/
│ │ ├── revenuecat.py # RevenueCat API client
│ │ └── adapty.py # Adapty API client
│ ├── auth/
│ │ └── credentials.py # Credential management
│ ├── models/
│ │ └── revenuecat.py # Data models
│ ├── tools/ # Migration tools
│ ├── utils/
│ │ └── logging.py # Logging utilities
│ └── server.py # MCP server implementation
├── tests/ # Test files
├── pyproject.toml # Project configuration
└── README.md # This file
Running Tests
# Run all tests
pytest
# Run specific test
pytest tests/test_revenuecat_export.py
# Run with coverage
pytest --cov=src/rc_adapty_migration
Testing the MCP
-
Test RevenueCat export:
python test_revenuecat_full_chain.py
-
Test Adapty migration:
python test_adapty_full_chain.py
Security Considerations
- RevenueCat API Key: Stored in environment variables
- Adapty Credentials: Provided at runtime, never stored
- Network: Use HTTPS for all API communications
- Validation: Validate all user inputs and API responses
- Logging: No sensitive credentials are logged
Troubleshooting
Common Issues
-
"API key not found"
- Ensure
REVENUECAT_API_KEY
environment variable is set - Verify the API key is valid and has proper permissions
- Ensure
-
"Login failed"
- Check Adapty credentials
- Ensure the account has proper permissions
-
"No projects found"
- Verify RevenueCat account has projects
- Check API key permissions
Debug Mode
Enable debug logging:
export LOG_LEVEL=DEBUG
rc-adapty-migration-mcp
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests
- Submit a pull request
License
MIT License - see LICENSE file for details
Support
For issues and questions:
- Create an issue on GitHub
- Check the troubleshooting section
- Review the test files for examples
Roadmap
- [ ] Enhanced validation tools
- [ ] Migration rollback functionality
- [ ] Advanced analytics and reporting
- [ ] Multi-project migration support
- [ ] Custom field mapping
- [ ] Migration templates
推荐服务器

Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。