Mnemex runs on your own machine — the client starts it, so there is no endpoint to ping. 14 installs a week from pypi. Last commit 17 Feb 2026.
Temporal memory for AI with decay and reinforcement. Two-layer storage (JSONL + Markdown).
Every tool that appeared, vanished or quietly changed what it asks for. Recorded since 9 August 2026. No other catalogue keeps this.
We read the source, 4 d ago · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
<div class="highlight"><pre><span></span><code><a id="__codelineno-17-1" name="__codelineno-17-1" href="#__codelineno-17-1"></a><span class="c1"># crontab -e</span>
этот файл ставится пользователю, но в репозитории его нет
- Two-phase pipeline: Extraction → Update - 26% accuracy boost over OpenAI's memory feature - 91% lower latency vs. full-context approach
result = subprocess.run(
1. IMDMR: Multi-Dimensional Memory Retrieval (Nov 2025)
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add mnemex -- uvx mnemex
{
"mcpServers": {
"mnemex": {
"command": "uvx",
"args": [
"mnemex"
]
}
}
}
[mcp_servers.mnemex]
command = "uvx"
args = ["mnemex"]
{
"mcpServers": {
"mnemex": {
"command": "uvx",
"args": [
"mnemex"
]
}
}
}
{
"servers": {
"mnemex": {
"type": "stdio",
"command": "uvx",
"args": [
"mnemex"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
MNEMEX_STORAGE_PATH (Storage path for short-term memory JSONL files), LTM_VAULT_PATH (Path to Obsidian vault for long-term memory), MNEMEX_DECAY_MODEL (Decay model: power_law (default), exponential, or two_component), MNEMEX_PL_HALFLIFE_DAYS (Half-life in days for power-law decay model).
Local persistent memory for AI agents with recall, search, forgetting, and encrypted storage.
Local persistent memory for AI agents with recall, search, forgetting, and encrypted storage.
Local-first, auditable memory for MCP-compatible AI agents with explicit writes and SQLite storage.
Durable, revision-pinned memory storage and retrieval for AI conversations.
AI assistant with RAG, web search, filesystem and memory. MCP sub-agent for Claude Code.
Long-term memory for AI assistants. Isolated per-user storage, recall across conversations.
Search and read arXiv papers, with local paper storage.
MCP server for the Longhorn distributed storage manager API, with a --read-only mode.
Answers built from our own checks of this server.