Amplify Doc MCP

🚀 AWS Amplify MCP Server Documentation

This is a documentation for the AWS Amplify MCP server, which provides search and query services for Amplify documentation.

🚀 Quick Start

The AWS Amplify MCP server is a powerful tool for searching and querying Amplify documentation. It supports advanced search syntax and intelligent sorting to provide accurate and relevant results.

📦 Installation

There is no specific installation process provided in the original document. If you want to run this server, you need to ensure that your system meets the following requirements:

• Node.js 18.x or higher (tested up to Node.js 20.18.2)
• npm 8.x or higher (tested up to npm 10.8.2)

✨ Features

• Advanced Query Parsing: The server can parse user search requests and recognize advanced syntax such as boolean operators and wildcards.
• Directory Optimization: Optimize the file scope to be searched based on keywords and the selected Amplify version.
• Intelligent Sorting: Sort search results according to multiple factors to ensure high relevance.
• Content Extraction: Extract and format relevant content from matching files.
• Caching Mechanism: Cache search results to improve performance for repeated queries.

💻 Usage Examples

Basic Usage

The configuration file config.json is used to configure the server. Here is an example:

{
  "awsRegion": "cn-north-1",
  "awsProfile": "default",
  "amplifyGeneration": "gen2",
  "autoUpdateInterval": 60,
  "maxCacheAge": 86400,
  "indexPath": "/var/www/html/amplify-docs-search"
}

Advanced Usage

The server has a series of query processing steps, including query parsing, directory optimization, intelligent sorting, content extraction, and caching. Here is a detailed description of the query processing flow:

1. Query Parsing: The server parses the user's search request and recognizes advanced syntax such as boolean operators and wildcards.
2. Directory Optimization: Optimize the file scope to be searched based on keywords and the selected Amplify version (Gen 1, Gen 2, or both).
3. Intelligent Sorting: Results are sorted according to the following factors:
   • Whether a specific version (Gen 1 or Gen 2) is mentioned in the query.
   • Whether the query involves settings, CLI commands, or resource creation.
   • The relevance of the document to the query context.
   • Exact matches in the title.
   • The number of matches and the importance of the document.
4. Content Extraction: Extract relevant content from each matching file and format it.
5. Caching Mechanism: Search results are cached to improve performance for repeated queries.

📚 Documentation

Project Structure

The project has the following structure:

• src/index.ts: Main server implementation.
• src/config.ts: Configuration loading and processing.
• src/git.ts: Git repository management.
• src/cache.ts: Search result caching.
• src/directory.ts: Directory structure management.
• src/types/: TypeScript type definitions.
• scripts/build.js: Documentation build script.
• bin/mcp: Executable script for running the server.

Usage Tips

⚠️ Important Note

For optimal search results, it is recommended to set "amplifyGeneration" to either "gen1" or "gen2".

💡 Usage Tip

• Use specific technical terms instead of general phrases.
• Include category names to narrow down the results (e.g., use "Storage owner access" instead of just "Access").
• Use quotes for exact phrase matching.
• Include abbreviations and alternative words to improve search results.
• Set an appropriate autoUpdateInterval according to your needs (a higher value can reduce server load).
• Utilize the caching mechanism for frequently accessed queries.
• Consider using filesOnly: true for an initial broad search to improve performance.

🔧 Technical Details

The server uses a series of algorithms and techniques to achieve efficient search and query functions, including query parsing, directory optimization, and intelligent sorting. These techniques ensure that users can get accurate and relevant search results.

📄 License

This project is open - source under the MIT License.