### Install Membase MCP Server (Shell)

Source: https://github.com/unibaseio/membase-mcp/blob/main/README.md

Instructions for cloning the membase-mcp repository and running the server using uv. This setup allows for the decentralized storage and retrieval of AI agent memories.

```shell
git clone https://github.com/unibaseio/membase-mcp.git
cd membase-mcp
uv run src/membase_mcp/server.py
```

--------------------------------

### Configure Membase MCP Server (JSON)

Source: https://github.com/unibaseio/membase-mcp/blob/main/README.md

Example JSON configuration for integrating the Membase MCP Server with AI frameworks like Claude, Windsurf, Cursor, or Cline. It specifies server command, arguments, and environment variables for account, conversation ID, and instance ID.

```json
{
  "mcpServers": {
    "membase": {
      "command": "uv",
      "args": [
        "--directory",
        "path/to/membase-mcp",
        "run", 
        "src/membase_mcp/server.py"
      ],
      "env": {
        "MEMBASE_ACCOUNT": "your account, 0x...",
        "MEMBASE_CONVERSATION_ID": "your conversation id, should be unique",
        "MEMBASE_ID": "your sub account, any string"
      }
    }
  }
}
```

--------------------------------

### Membase MCP Server API Functions (APIDOC)

Source: https://github.com/unibaseio/membase-mcp/blob/main/README.md

Core functions for interacting with the Membase MCP Server to manage AI agent conversations and memories. These functions allow for retrieving conversation IDs, switching between conversations, saving new messages, and fetching historical messages.

```APIDOC
Membase MCP Server API:

- get_conversation_id()
  - Purpose: Retrieves the unique identifier for the current AI agent conversation.
  - Returns: The current conversation ID (string).

- switch_conversation(conversation_id: str)
  - Purpose: Switches the server's context to a different, specified conversation.
  - Parameters:
    - conversation_id: The ID of the conversation to switch to (string).
  - Returns: Confirmation of the switch or an error if the conversation ID is invalid.

- save_message(message: dict)
  - Purpose: Persists a new message or memory into the currently active conversation.
  - Parameters:
    - message: A dictionary containing the message content and associated metadata.
  - Returns: Confirmation of successful save or an error message.

- get_messages(n: int)
  - Purpose: Retrieves the last 'n' messages from the current conversation's history.
  - Parameters:
    - n: The number of recent messages to fetch (integer).
  - Returns: A list of message dictionaries, ordered chronologically.
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.