Memphora

Connect your assistant to Memphora to remember, search, and recall information across conversations.

python

1

GitHub Stars

python

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "memphora-memphora-mcp": {
      "command": "uvx",
      "args": [
        "memphora-mcp"
      ],
      "env": {
        "MEMPHORA_API_KEY": "YOUR_API_KEY",
        "MEMPHORA_USER_ID": "YOUR_UNIQUE_USER_ID"
      }
    }
  }
}

Memphora MCP Server connects your AI assistants to Memphora, enabling memory across conversations, knowledge-base search, automatic context extraction, and recall of preferences and facts to enhance your interactions.

How to use

You integrate Memphora MCP Server with your MCP client to give your assistant memory capabilities. You can store memories, search your memories when needed, and have your assistant automatically reference relevant past conversations to improve responses. The server provides tools to store, search, extract, and list memories, making it easier to remember who you are, what you like, and what you’ve discussed.

How to install

Prerequisites you need before installing: Python and pip installed on your system. If you plan to use the uvx runtime, ensure uvx is available on your machine.

pip install memphora-mcp

Alternatively, you can run the MCP server with uvx (recommended for Claude Desktop users):

uvx memphora-mcp

Configuration and usage notes

To connect your MCP client to Memphora, provide the API key and a user identifier used to scope memories to your account. The client examples below show how to wire Memphora into different MCP clients.

Additional content you can configure with Memphora MCP

Configure your MCP client to point to Memphora using the following example configurations. Each configuration enables the Memphora MCP server as a local, real-time bridge between your assistant and Memphora.

Development notes

When you run the server locally, you set your API key in the environment so the client can authenticate with Memphora. The API key remains secret on your machine and is not exposed to the client beyond the configured runtime.

Security and privacy considerations

Memphora stores memories in Memphora’s cloud with isolated storage per user. All communication between your client and Memphora is encrypted over HTTPS. Your API key remains on your device, and access to memories is scoped to your user.

Available tools

memphora_search

Search memories for relevant information across your stored memories.

memphora_store

Store new information so you can recall it in future conversations.

memphora_extract_conversation

Automatically extract memories from a completed conversation for later recall.

memphora_list_memories

List all memories that have been stored for your account.

memphora_delete

Delete a specific memory from your storage.