Loading...
Membase-MCP server provides decentralized, verifiable storage for AI agent memories via the Membase protocol, enabling persistent conversation history and knowledge retrieval.
Boost this tool
Subscribe to listing upgrades or segmented pushes.
Membase-MCP server provides decentralized, verifiable storage for AI agent memories via the Membase protocol, enabling persistent conversation history and knowledge retrieval.
The Membase-MCP server offers a moderate level of safety. While it provides a limited set of functions and stores data on a decentralized network, the reliance on environment variables for authentication and potential lack of input sanitization pose risks. It's safer when used with strong environment variable management and input validation.
Performance will be affected by the decentralized storage network latency. Consider the number of messages retrieved and the frequency of write operations to optimize performance.
Cost will depend on the usage of the Unibase DA network for storage. Consider the size and number of messages stored, as well as any fees associated with the network.
{
"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"
}
}
}
}get_conversation_idRetrieves the current conversation ID.
Read-only operation, no side effects.
switch_conversationSwitches to a different conversation using its ID.
Changes the current conversation context, potentially affecting subsequent operations.
save_messageSaves a message/memory into the current conversation.
Writes data to the decentralized storage, but is limited to message content.
get_messagesRetrieves the last n messages from the current conversation.
Read-only operation, retrieves existing messages.
Environment Variable
cloud
The Membase-MCP server offers a moderate level of safety. While it provides a limited set of functions and stores data on a decentralized network, the reliance on environment variables for authentication and potential lack of input sanitization pose risks. It's safer when used with strong environment variable management and input validation.
Autonomy is limited by the exposed functions. The agent can read and write messages within a conversation, but cannot perform administrative tasks or access sensitive system resources. Sandboxing is achieved through the limited function set and the decentralized storage.
Production Tip
Ensure proper management of MEMBASE_ACCOUNT and MEMBASE_CONVERSATION_ID environment variables to prevent unauthorized access and data leakage.
Membase is a decentralized memory layer for AI agents powered by Unibase, providing secure and persistent storage.
Clone the repository, navigate to the directory, and run `uv run src/membase_mcp/server.py`.
MEMBASE_ACCOUNT, MEMBASE_CONVERSATION_ID, and MEMBASE_ID are required.
Use the `switch_conversation` tool with the desired conversation ID.
Messages are stored on the Unibase DA network for decentralized, verifiable storage.
Data is stored on a decentralized network, but security depends on proper management of your MEMBASE_ACCOUNT and the security of the Unibase network.
The provided documentation does not explicitly mention a delete function. It is likely that messages cannot be deleted through this MCP server.