Advertise
MCP.so
Sign In

Memanto MCP Server

@moorcheh-ai

Repository

Basic information

Category

Other

Transports

stdio

Author

moorcheh-ai

Submitted by

Hetkumar Patel

About Memanto MCP Server

MEMANTO is a memory agent. It remembers, recalls, and answers - so your agents can achieve long-term goals and avoid confusion.

#memory#ai-agents#long-term-memory#rag#semantic-memory#llm-memory#agent-memory#moorcheh#stateful-ai#memanto

Memanto MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "memanto": {
      "command": "memanto-mcp",
      "env": {
        "MOORCHEH_API_KEY": "<your-key>",
        "MEMANTO_DEFAULT_AGENT_ID": "my-project"
      }
    }
  }
}

Memanto MCP Server tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Memanto MCP?

Memanto MCP is a server that exposes Memanto's persistent semantic memory primitives — remember, recall, answer, and related tools — as Model Context Protocol (MCP) tools. It allows any MCP-compatible client (Claude Desktop, Cursor, Windsurf, Cline, Continue, Goose, etc.) to store and retrieve typed semantic memories in a shared namespace using a Moorcheh API key. The server requires Python 3.10+ and a Moorcheh API key (free tier: 100K operations per month).

How to use Memanto MCP?

Install with pip install memanto-mcp. Set the MOORCHEH_API_KEY environment variable and optionally MEMANTO_DEFAULT_AGENT_ID. Add the server to your MCP client's configuration file (e.g., claude_desktop_config.json). The server registers seven memory tools by default; four agent‑management tools can be enabled via MEMANTO_EXPOSE_ADMIN=true. Transport can be stdio, sse, or streamable‑http, configured via environment variables or CLI flags (memanto-mcp --transport sse --port 8765).

Key features of Memanto MCP

  • Persistent semantic memory across MCP-compatible agents
  • Sub-90 ms retrieval with conflict detection
  • Shared memory namespace for multiple agents
  • Supports 13 memory types and 6 provenance values
  • RAG-based grounded answer tool using memories
  • Point-in-time and differential recall capabilities

Use cases of Memanto MCP

  • Remember user preferences and instructions across chat sessions
  • Recall decisions and context from previous conversations
  • Perform RAG over an agent’s stored knowledge for grounded answers
  • Track goals, commitments, and observations over time
  • Batch import extracted facts from documents using batch_remember

Frequently asked questions

1

What dependencies does Memanto MCP require?

Python 3.10+ and a Moorcheh API key. The free tier provides 100K operations per month.

2

How is the server authenticated?

The server authenticates to Moorcheh via the API key but does **not** authenticate inbound MCP clients; production deployments should pair it with a reverse proxy and authentication.

3

Which transports are supported?

The server supports `stdio` (default), `sse` (Server-Sent Events), and `streamable-http` (recommended modern transport). Transport is configured via the `MEMANTO_MCP_TRANSPORT` environment variable or CLI flags.

4

How does session management work?

On first memory tool call, a JWT session is activated and auto‑renews before expiry. Sessions are intentionally kept alive on shutdown so other Memanto clients can share them.

5

Can I embed Memanto MCP programmatically?

Yes. You can construct the FastMCP instance using `MCPServerSettings` and `build_server` from the `memanto_mcp` package, then add your own tools and run the server.

Comments

More Other MCP servers

MCP Toolbox for Databases

googleapis

MCP Toolbox for Databases is an open source MCP server for databases.

15.9K

Figs, Relationship Expert

Figs O'Sullivan, LMFT

Couples therapist Figs O'Sullivan (LMFT, 17 years, endorsed by Sue Johnson, creator of EFT) inside your AI. Ask relationship questions answered from his clinical writing, name the pattern in a fight, get a three-step rep

Blender

ahujasid

Open-source MCP to use Blender with any LLM

23.6K

ICSS

chokcoco

不止于 CSS

20.7K

Awesome Mlops

visenger

A curated list of references for MLOps

14.0K
Advertise