Memory Industry runs on your own machine — the client starts it, so there is no endpoint to ping. 612 installs a week from pypi. Last commit 28 Oct 2026.
MemoryIndustry persistent memory MCP server: BM25+MMR retrieval and audit.
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.
пакет cuba-memorys ссылается на leandropg19/cuba-memorys
const out = execFileSync(bin, ["--version"], {
const localBin = path.join(__dirname, ".bin", binName);
SELECT r.from_entity, r.to_entity,
r.strength / LN(1.0 + COALESCE(deg.degree, 1)) AS dampened_weight
$("checks").innerHTML = table(["", "Chequeo", "Detalle"], rows);
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 memory-industry -e 'DATABASE_URL=<DATABASE_URL>' -- npx -y memory-industry
{
"mcpServers": {
"memory-industry": {
"command": "npx",
"args": [
"-y",
"memory-industry"
],
"env": {
"DATABASE_URL": "<DATABASE_URL>"
}
}
}
}
[mcp_servers.memory-industry]
command = "npx"
args = ["-y", "memory-industry"]
env = { DATABASE_URL = "<DATABASE_URL>" }
{
"mcpServers": {
"memory-industry": {
"command": "npx",
"args": [
"-y",
"memory-industry"
],
"env": {
"DATABASE_URL": "<DATABASE_URL>"
}
}
}
}
{
"servers": {
"memory-industry": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"memory-industry"
],
"env": {
"DATABASE_URL": "<DATABASE_URL>"
}
}
}
}
Put your own value in place of
<DATABASE_URL> (PostgreSQL connection URL (e.g. postgresql://user:pass@localhost:5432/brain)).
The author’s registry entry marks it as required.
Optional settings from the registry entry, left out of the block above:
CUBA_PROJECT_FILTER (Set to 'off' to disable v0.8 project scoping filters (admin/debug only)), CUBA_SYNC_DIR (Directory used by cuba_sync export/import (default ./.memory-industry/, falls back to ./.cuba-memorys/)), CUBA_JUDGE (cuba_juez backend selector: nli | mcp_sampling | claude_cli | heuristic | auto (default auto)), CUBA_JUEZ_CLI (Subprocess CLI used by ClaudeCodeJudge (default 'claude')), CUBA_JUEZ_MODEL (Model name passed to the judge backend (default 'claude-haiku-4-5')), CUBA_REM_RELATION_BATCH (Isolated entities the REM cycle scans for relations per run (default 5, 0 disables)), CUBA_RERANKER_PATH (Directory holding the bge-reranker ONNX model; enables cross-encoder reranking (+93% nDCG on GPU)), CUBA_RERANK_FIXED_SHAPE (Pad reranker batches to a constant 512 tokens. Defaults on for GPU builds, off on CPU).
Persistent memory MCP server. 28 tools, BM25+MMR+OOD retrieval, tamper-evident audit, code graph.
SQLite-backed MCP server for persistent memory, full-text retrieval, and graph traversal.
Local-first MCP memory server for persistent LLM knowledge with BM25 retrieval.
Memori MCP server — persistent AI memory with recall and augmentation tools
Persistent memory MCP server for Claude Code, Cursor, and GitHub Copilot.
MCP server for persistent, searchable memory for AI agents.
Local-first persistent memory MCP server for AI coding agents.
MCP server for Memanto - persistent semantic memory for any MCP-compatible agent
Answers built from our own checks of this server.