Mcp External Memory

io.github.daedalusPublished by GitHub account daedalusv0.1.2Updated Oct 24, 2026

An MCP server that gives LLMs persistent, searchable semantic memory

The publisher's own server: Memory.fit

VerifiedSTDIODesktop onlyKnowledge & Memory

Overview

AI-generated overview

Gives an assistant persistent, searchable semantic memory it can store, search, update, and delete across sessions.

What you can do
This local MCP server exposes a small memory store as tools: memory_store persists text with an optional namespace, tags, and metadata; memory_search runs semantic search (cosine similarity) over stored memories; memory_get, memory_update, and memory_delete work on individual entries by ID; memory_list lists memories with namespace or tag filters and pagination; memory_stats reports counts, namespaces, and the database path. Embeddings come from a pluggable backend, with TF-IDF as the default and OpenAI or Ollama as alternatives.
When to use it
Worth adding when you want an assistant to remember facts, preferences, or notes between conversations and retrieve them by meaning rather than exact keywords. Useful for per-user or per-project namespaces, such as remembering a user's settings, and for keeping a local, inspectable memory store instead of relying on the model's context window.
Requirements
Runs as a local process over stdio, installed from PyPI as mcp-external-memory (pip install mcp-external-memory), so a Python environment is needed. No authentication, environment variables, or headers are declared in the manifest. The embedding backend is selected with the MEMORY_EMBED_BACKEND environment variable; the default TF-IDF backend needs no external service, while the OpenAI backend requires an API key and network access and Ollama requires a local Ollama instance.
Before you install
The server writes, updates, and deletes stored memories, so deletions are permanent and stored content persists on disk. Memories may contain personal or sensitive information, and choosing the OpenAI embedding backend sends memory text to a third party, so prefer the local TF-IDF or Ollama backends for sensitive data. The OpenAI backend also needs its own API credentials, which the manifest does not declare.

Installation

In SourceWeft

  1. Open Mcp External Memory in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

mcp-external-memory

An MCP server that gives LLMs persistent, searchable semantic memory.

[PyPI] [Python] [Coverage] [Ruff]

Install

bash
pip install mcp-external-memory

Usage

python
from mcp_external_memory import memory_store, memory_search
# Store a memoryresult = memory_store(content="Alice prefers dark mode", namespace="users", tags=["alice", "ui"])
# Search memoriesresults = memory_search(query="what does Alice prefer?", namespace="users")

CLI

bash
mcp-external-memory --help

API

Tools

ToolDescription
memory_storePersist text + optional namespace/tags/metadata
memory_searchSemantic search (cosine similarity) over all memories
memory_getRetrieve a single memory by ID
memory_deleteDelete a memory by ID
memory_listList memories with optional namespace/tag filter + pagination
memory_statsCount of memories, namespaces, DB path
memory_updateUpdate an existing memory

Embedding Backends

The server supports multiple embedding backends:

  • TF-IDF (default): Pure Python, no external dependencies
  • OpenAI: Uses text-embedding-3-small model
  • Ollama: Local embeddings with Ollama

Set via MEMORY_EMBED_BACKEND environment variable.

Development

bash
git clone https://github.com/daedalus/mcp-external-memory.gitcd mcp-external-memorypip install -e ".[test]"
# run testspytest
# formatruff format src/ tests/
# lintruff check src/ tests/
# type checkmypy src/

MCP Registry

mcp-name: io.github.daedalus/mcp-external-memory

Source: README.md at commit ad14722

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.1.2LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

Other Memory integrations

Memory.fit

fit.memoryOfficial publisher

Hosted Memory.fit server that lets an AI read your training, sleep, recovery and nutrition data and write back weekly plans, targets and meal plans.

Streamable HTTPWeb executableAuth
Location & LifestyleCredential requiredOct 29, 2026

pg.ddx.io: the PostgreSQL community's memory

io.ddx.pgOfficial publisher

pg.ddx.io: the PostgreSQL community's memory's hosted MCP server, published by pg.ddx.io; no sign-in required; Streamable HTTP; 29 tools.

Streamable HTTPWeb executable
Databases29 toolsOct 24, 2026

var.gg Naming Memory

gg.varOfficial publisher

Remote MCP server that suggests and searches code identifier names from domain concepts in Korean, Japanese, Chinese, Vietnamese and English, and exposes ERD…

Streamable HTTPWeb executable
Databases7 toolsOct 27, 2026

Counterparts

ai.counterparts

Local-first long-term memory for Claude Code and Claude Desktop: the assistant writes notes about you and your projects and recalls them in later sessions.

STDIODesktop only
Knowledge & MemoryNot validated yetNov 7, 2026

LAHZE - Persian AI Memory Platform

Published by GitHub account pedramgholizadeh

LAHZE - Persian AI Memory Platform's hosted MCP server; no sign-in required; Streamable HTTP; 13 tools.

Streamable HTTPWeb executable
Location & Lifestyle13 toolsNov 3, 2026