Ibproduct Ib MCP Cache Server

🚀 Memory Cache Server

A Model Context Protocol (MCP) server that reduces token consumption by efficiently caching data between language model interactions. It is compatible with any MCP client and any language model that uses tokens.

🚀 Quick Start

📦 Installation

1. Clone the repository:

git clone git@github.com:ibproduct/ib-mcp-cache-server
cd ib-mcp-cache-server

2. Install dependencies:

npm install

3. Build the project:

npm run build

4. Add it to your MCP client settings:

{
  "mcpServers": {
    "memory-cache": {
      "command": "node",
      "args": ["/path/to/ib-mcp-cache-server/build/index.js"]
    }
  }
}

5. The server will start automatically when the MCP client is enabled.

✅ Verification

When the server is running properly, you will see the following in the terminal:

1. The message: "Memory cache MCP server is running on stdio".
2. Performance improvement when accessing the same data multiple times.
3. Faster responses, especially for repeated operations.

📚 Documentation

⚙️ Configuration

All configuration parameters are as follows:

💡 How to Reduce Token Consumption

The server reduces token consumption in the following ways:

1. Automatic Caching: All responses are automatically cached unless the Cache-Control: no-cache header is set.
2. Expiration Mechanism: Cache entries will automatically expire after the set time.
3. Hit Rate Statistics: The server tracks the number of cache hits and displays it in the console.

🗄️ Automatic Cache Management

1. The server loads the configuration file and initializes the cache when it starts.
2. It regularly checks the cache space usage and automatically cleans up when the limit is exceeded.
3. It supports setting the maximum memory usage to avoid affecting system performance.

🛠️ Optimization Tips

• Adjust the number of entries: Increase MAX_ENTRIES according to project requirements to improve the hit rate.
• Extend the validity period: Appropriately increase DEFAULT_TTL to reduce the number of cache invalidations.
• Monitor usage: Regularly check the hit rate and cache size to ensure optimal performance.

🌐 Environment Variable Configuration

You can set environment variables in the MCP client to override the default values:

{
  "mcpServers": {
    "memory-cache": {
      "command": "node",
      "args": ["/path/to/build/index.js"],
      "env": {
        "MAX_ENTRIES": "5000",
        "MAX_MEMORY": "209715200",  // 200MB
        "DEFAULT_TTL": "7200",      // 2 hours
        "CHECK_INTERVAL": "120000",  // 2 minutes
        "STATS_INTERVAL": "60000"    // 1 minute
      }
    }
  }
}

You can also specify a custom configuration file path:

{
  "env": {
    "CONFIG_PATH": "/path/to/your/config.json"
  }
}

The server will:

1. Look for the default configuration file in its directory.
2. Apply any environment variable overrides.
3. Use the default values if not specified.

🧪 Testing Cache Effectiveness

To observe the actual effect of the cache, try the following scenarios:

📃 File Reading Test

• Read and analyze a large file.
• Ask the same question again.
• The second response should be faster because the file content is cached.

📊 Data Analysis Test

• Perform some data analysis.
• Request the same analysis again.
• The second analysis will use the cached results.

📂 Project Navigation Test

• Explore the project structure.
• Query the same file/directory again.
• The directory list and file content will be provided from the cache.

When the cache is working properly, you will notice:

• Faster responses for repeated operations.
• Consistent answers for unchanged content.
• No need to re-read unchanged files.