membase-mcpThis server lets AI assistants save messages and recall old chats instead of starting fresh every time. You can switch between topics, store important notes, and ask the AI to look back at earlier par

— (0)

Reviewed by Daniel Nikulshyn·Updated July 2026

Overview

Membase-MCP is a server that provides a decentralized memory layer for AI agents, enabling them to save and recall conversation history. It allows AI assistants to switch between topics, store important notes, and retrieve earlier parts of a conversation. The server integrates with the Membase protocol for secure, persistent storage of conversation history and interaction records. This ensures agent continuity, personalization, and traceability. The Membase-MCP Server can be installed and run locally, and it provides functions for managing conversations and messages.

Key features

get_conversation_id
switch_conversation
save_message
get_messages

Pricing

Model: Free
Category: MCP Servers
Rating: No reviews yet

Use cases

AI Assistant Conversation Continuity

An AI assistant uses Membase-MCP to save and recall conversation history, allowing it to pick up where it left off with a user.

Knowledge Management

A knowledge management system utilizes Membase-MCP to store and retrieve interaction records and conversation history, ensuring traceability and personalization.

Pros & Cons

Pros

enables AI agents to recall old chats
allows switching between topics
provides secure and persistent storage
ensures agent continuity and personalization

Cons

requires installation and configuration
dependent on Membase protocol and Unibase DA network
limited documentation and community support

Reviews

No reviews yet. Be the first!

Q&A

What are the main limitations of using Membase-MCP?

It requires manual installation and configuration, depends on the Membase protocol and Unibase DA network, and has limited documentation and community support, which may increase the learning curve for new users.

Asked by Farah Rahimi · Aug 19, 2025

Does Membase-MCP integrate with other AI platforms?

Yes, it can be integrated with Claude, Windsurf, Cursor, and Cline by configuring the `mcpServers` section with the server command, arguments, and required environment variables, enabling those agents to upload and retrieve memory from the Unibase DA network.

Asked by Hasan Demir · Aug 3, 2025

What functions does the server expose for managing conversations?

The server provides four main functions: `get_conversation_id` to retrieve the current ID, `switch_conversation` to change topics, `save_message` to store a message or note, and `get_messages` to fetch the last n messages from the current conversation.

Asked by Dumisani Ndlovu · Jul 6, 2025

How do I install and run Membase-MCP locally?

Clone the repo with `git clone https://github.com/unibaseio/membase-mcp.git`, navigate into the folder, then start the server using `uv run src/membase_mcp/server.py`. You must set environment variables MEMBASE_ACCOUNT, MEMBASE_CONVERSATION_ID, and MEMBASE_ID before running.

Asked by Hannah Goldberg · Jun 9, 2025