Implementing Persistent Memory Using a Local Knowledge Graph in Claude Desktop

The Memory Memory Clated Claude Desktop allows to remember and organize information about a user via multiple conversations. It can store things like user preferences, previous conversations and personal details. Since the information is saved as a graphic fee for knowledge, Claude can understand the relationships between different parts of the information. This leads to more custom responses and reduces repetition – you will not have to explain the same things over and over again.

In this tutorial, we will implement a simple fixed memory using a local graph for the Claude desktop, to help him remember the user information via chats and provide more customized and consistent responses.

Install node.js

We will use NPX to run a knowledge diagram for knowledge, and for this, node.js.

• Download the latest Release From Node.js from Nodejs.org
• Stabilize.
• Leave all settings by default and complete the installation

Claude desktop designation

You can download the latest version of the Claude desktop on https://claud.AI/download. Next, you will need to configure CLADE to communicate with your MCP server. To do this, open the Claude_Desktop_config.json file in the Claude guide using any text editor. If the file is not present, then continue to create it manually.

In McP.json file, enter the following code:

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

Now, we need to configure Claude in order to be able to use the knowledge chart to create entities, build relationships and recover relevant information.

• Go to File> Settings> Claude settings> configuration.

• In the Personal Preferences Department, add the following text:

(This preference will be applied automatically to all conversations.)

Follow these steps for each interaction:

1. User Identification:
   - You should assume that you are interacting with default_user
   - If you have not identified default_user, proactively try to do so.

2. Memory Retrieval:
   - Always begin your chat by saying only "Remembering..." and retrieve all relevant information from your knowledge graph
   - Always refer to your knowledge graph as your "memory"

3. Memory
   - While conversing with the user, be attentive to any new information that falls into these categories:
     a) Basic Identity (age, gender, location, job title, education level, etc.)
     b) Behaviors (interests, habits, etc.)
     c) Preferences (communication style, preferred language, etc.)
     d) Goals (goals, targets, aspirations, etc.)
     e) Relationships (personal and professional relationships up to 3 degrees of separation)

4. Memory Update:
   - If any new information was gathered during the interaction, update your memory as follows:
     a) Create entities for recurring organizations, people, and significant events
     b) Connect them to the current entities using relations
     b) Store facts about them as observations

Once everything is formed, you will see 9 MCP tools available for the knowledge graph server. These tools allow you: create entities, create relationships, add notes, delete entities, delete notes, delete relationships, read the graph, search contract, and open contract.

In addition, the text that we added in the Claude preferences allows these tools to use automatically during conversations.

Even if we go to a new chat, Claude will remember information from the previous chats via the knowledge chart. This MCP tool integrates Claude’s ability to create, modify and use knowledge, making it a strong assistant such as database management and SQL With this memory system, Claude becomes a more intelligent, response and consistent tool for all your future reactions. For more details about the knowledge memory server, you can visit this link, as you will find resources to help you build more advanced applications.

Also, do not forget to follow us twitter And join us Telegram channel and LinkedIn GrOup. Don’t forget to join 90k+ ml subreddit.

🔥 [Register Now] The virtual Minicon Conference on Agency AI: Free Registration + attendance Certificate + 4 hours short (May 21, 9 am- Pacific time)

I am a graduate of civil engineering (2022) from Islamic Melia, New Delhi, and I have a strong interest in data science, especially nervous networks and their application in various fields.