Memories With Lessons MCP Server

🚀 Knowledge Graph Server

The Knowledge Graph Server provides a comprehensive set of functions for managing knowledge graphs, including entity management, relationship establishment, and error handling course management.

✨ Features

Knowledge Graph Operations

• create_entity: Create a new entity.
• get_entity: Retrieve information about a single entity.
• update_entity: Update the information of an existing entity.
• delete_entity: Delete a specified entity.
• link_entities: Establish a relationship between two entities.
• unlink_entities: Remove the relationship between two entities.
• query_graph: Return matching entities and relationships based on query conditions.
• add_observation: Add an observational fact to the knowledge graph.
• get_observations: Retrieve all observational facts for a specific entity.
• update_observation: Update the content of an existing observational fact.
• delete_observation: Delete a specified observational fact.

New Tools

• create_lesson

  • Create a new lesson using errors and their solutions.
  • Input: lesson (object)
    • Contains error patterns, solution steps, and metadata.
    • Automatically tracks creation time and updates.
    • Validates the completeness of solution steps.

• find_similar_errors

  • Find similar errors and their solutions.
  • Input: errorPattern (object)
    • Contains error type, message, and context.
    • Returns matching lessons sorted by success rate.
    • Includes relevant solutions and validation steps.

• update_lesson_success

  • Update the success tracking information of a lesson.
  • Input:
    • lessonName (string): The name of the lesson to update.
    • success (boolean): Indicates whether the solution is effective.
  • Updates success rate and frequency metrics.

• get_lesson_recommendations

  • Get relevant lesson recommendations based on context.
  • Input: context (string)
  • Returns lessons sorted by relevance and success rate.
  • Includes complete solution details and validation steps.

🚀 Quick Start

Docker Configuration

Add the following content to claude_desktop_config.json:

{
  "mcpServers": {
    "memory": {
      "command": "docker",
      "args": ["run", "-i", "-v", "claude-memory:/app/dist", "--rm", "mcp/memory"]
    }
  }
}

NPX Configuration

{
  "mcpServers": {
    "memory": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-memory"
      ]
    }
  }
}

Custom Environment Variable Configuration

{
  "mcpServers": {
    "memory": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-memory"
      ],
      "env": {
        "MEMORY_FILE_PATH": "/path/to/custom/memory.json"
      }
    }
  }
}

• MEMORY_FILE_PATH: Specify the path to the memory storage JSON file (default is memory.json in the server directory).

System Prompt

Adjust the prompt according to the usage scenario. Here is an example for chat personalization configuration, which can be used in the "Custom Instructions" field of the Claude.ai project:

Follow these steps for each interaction:

1. User Identification:
   - Assume you are interacting with default_user.
   - If not, skip to the next step.

2. Context Management:
   - Use the memory graph to store context information.
   - Retrieve relevant context when needed.

3. Target Response:
   - Analyze user input.
   - Generate meaningful responses.

💻 Usage Examples

Knowledge Graph Example

{
  "entities": [
    {
      "id": "e1",
      "name": "Alice",
      "age": 30,
      "occupation": "Engineer"
    },
    {
      "id": "e2",
      "name": "Bob",
      "age": 28,
      " occupation": "Designer"
    }
  ],
  "relations": [
    {
      "id": "r1",
      "from": "e1",
      "to": "e2",
      "type": "colleagues"
    }
  ],
  "observations": [
    {
      "id": "o1",
      "entity": "e1",
      "content": "Alice works at Google."
    },
    {
      "id": "o2",
      "entity": "e2", 
      "content": "Bob works at Microsoft."
    }
  ]
}

Build Instructions

docker build -t mcp-memory-server .
docker run -p 3000:3000 mcp-memory-server

📄 License

This project is licensed under the MIT License. Please refer to the LICENSE file for details.

Thanks to @your_name for the contribution!