MCP SERVER · OFFICIAL PROJECTLong-term memory any MCP client can share — typed, semantically searchable, and answerable as of a date.
LABSCO SUMMARY
Most memory servers stop at write and semantic search. The two that earn their place here are recall_as_of, which reconstructs what was known at an earlier moment, and recall_changed_since, which catches an agent up on what moved while it was away. Writes that do not name a source are attributed to the connected client from the handshake, so two editors sharing one namespace stay distinguishable in recall with no extra configuration.
A memory server backed by the Moorcheh service. It stores typed memories against an agent namespace and retrieves them by semantic similarity, by recency, or as of a point in time.
A Moorcheh API key in MOORCHEH_API_KEY. Install memanto-mcp with pip; it needs Python 3.10 or newer. MEMANTO_DEFAULT_AGENT_ID names the memory namespace so calls can leave agent_id out, and that agent is created on first use. MEMANTO_MCP_TRANSPORT switches between stdio, sse and streamable-http, with MEMANTO_MCP_HOST and MEMANTO_MCP_PORT for the network ones, and CLI flags override the environment. Over a network transport it authenticates upward to Moorcheh with your key but does not authenticate inbound clients — put a reverse proxy with auth in front of it before exposing the port.
One command plus a key — pip install memanto-mcp, then supply credentials
USE CASES
Store it once with its type, and search memory before asking again.
Extract them, then persist the whole list in one call rather than a hundred round trips.
Get a synthesized answer grounded in memory, with the sources it came from attached.
Pull whatever was created or changed since you last looked, rather than re-reading everything.
Recall only what was known before that date, so later knowledge does not contaminate the picture.
Read the newest memories first, then search for anything specific they raise.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
answerAsk a natural-language question and get an LLM-generated answer grounded ONLY in the agent's stored memories (RAG). Prefer this over recall when you need a synthesized answer rather than a ranked list. Returns the answer text plus the supporting memory sources.
batch_rememberStore many memories at once (up to 100). Use this when you have a list of independent facts to persist - e.g. extracting structured data from a document. For a single item, prefer remember.
recallSearch the agent's memories by semantic similarity. Returns the top-N most relevant items. Use this FIRST before asking the user to repeat information - the agent may already remember it. The query should be natural language ('what does the user prefer for code style?'), not keywords.
recall_as_ofPoint-in-time recall: return only memories that were known before the given timestamp. Use this when the user asks historical questions like 'what did we know on 2025-11-01?' or to reconstruct context at a previous moment.
recall_changed_sinceDifferential retrieval: return memories created or updated after the given timestamp. Use this for 'what's new since X?' or to catch up on activity between sessions.
recall_recentReturn the most recently stored memories (newest first). Use this to surface fresh context - e.g. 'what did we just decide?' - when you don't have a specific search query.
rememberStore a single piece of information in the agent's long-term memory. Use this whenever the user shares a stable fact, preference, decision, goal, or instruction you should recall in a future conversation. Memory is typed (13 categories) and carries confidence + provenance so later retrievals can rank and filter intelligently. Content is capped at 10000 chars - store atomic, self-contained statements.
BEFORE YOU INSTALL
7 tools when we connected; the project documents 11.
MOORCHEH_API_KEY in your environment
All 4 documented clients run the server on your machine.
INSTALLATION
pip install memanto-mcp{
"mcpServers": {
"memanto": {
"command": "memanto-mcp",
"env": {
"MOORCHEH_API_KEY": "<YOUR_MOORCHEH_API_KEY>",
"MEMANTO_DEFAULT_AGENT_ID": "my-assistant"
}
}
}
}{
"mcpServers": {
"memanto": {
"command": "memanto-mcp",
"env": {
"MOORCHEH_API_KEY": "<YOUR_MOORCHEH_API_KEY>",
"MEMANTO_DEFAULT_AGENT_ID": "cursor-workspace"
}
}
}
}MCP servers
fatmali
RetroRobAI
STiFLeR7
Moai-Team-LLC