MemoryGuard runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 30 Sep 2026.
Local-first governed MCP memory backend for coding agents.
What the code does
We read the source, 1 d ago · rules ccca72095570
A value the model can set ends up inside a file or shell call. That is not a flaw by itself — for a terminal server it is the job — but it is where things go wrong when it is not.
subprocess.Popen(
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.
completed = subprocess.run(
row = conn.execute(turn_select + f" JOIN conversation_sessions s ON s.session_id=t.session_id WHERE t.turn_id=? AND {where}", [turn_id, *args]).fetchone()
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
A tool parameter here reaches a dangerous call
That is not a flaw by itself — but it is where things go wrong when it is not the job. We re-read this code on every release. Watch it and you hear from us the day another one appears.
Alternatives to MemoryGuard
same job, measured the same wayLocal-first governed MCP memory backend for coding agents.
Local-first, governable long-term memory for AI coding agents
Local-first persistent memory MCP server for AI coding agents.
Local-first, cross-harness memory for AI coding agents.
Local-first persistent memory for coding agents and MCP clients.
Local-first persistent memory for AI coding agents.
Local-first, state-centric memory for coding agents. MCP server + CLI.
Local-first project memory and mistake-prevention guardrails for AI coding agents.
MemoryGuard — questions
Answers built from our own checks of this server.