superlocalmemory
Persistent local memory for AI agents. MCP server and plugins for Claude Code, Codex, Cursor, VS Code Copilot, Antigravity and more: your agents remember across
What is superlocalmemory?
Persistent local memory for AI agents. MCP server and plugins for Claude Code, Codex, Cursor, VS Code Copilot, Antigravity and more: your agents remember across sessions, on your machine. arXiv:2608.08253
Overview of superlocalmemory
What it is
SuperLocalMemory is a persistent local memory system for AI agents. It runs as an MCP server and provides plugins for Claude Code, Codex, Cursor, VS Code Copilot, Antigravity, and other tools. The memory is stored on the user's own machine and is designed to prevent agents from guessing when they do not have an answer. The project is described in arXiv:2608.08253.
Key features
- Memory is stored locally on the user's machine; no API key or cloud account is required for core remember and recall operations (Mode A).
- Agents can recall memories across sessions, including decisions, documents, screenshots, and PDFs.
- Memories are governed: the system tracks who saved what, when, and for which project, and can refuse to return a result when nothing relevant is stored.
- Supports importing history from ChatGPT, Claude, Mem0, Letta, or AGENTS.md/CLAUDE.md files, with the ability to undo the entire import in one click.
- Includes a per-prompt hook that checks memory before the agent answers, skipping greetings, arithmetic, and slash commands, and avoiding repetition of already-discussed memories.
- Cues allow memories to be triggered by file patterns, commands, error texts, repositories, or dates.
- Disputes are shown when two memories give conflicting values for the same thing, rather than silently overwriting.
Use cases
SuperLocalMemory is intended for developers and teams who use AI coding agents and want those agents to remember context across sessions without relying on cloud services. It can be used to store project decisions, configuration details, and documentation, and to recall them accurately. The system is also suitable for environments where data governance and local control are important.
Getting started
The easiest installation method is to copy the following prompt into an AI agent that can run commands:Install SuperLocalMemory on this computer for me. Followhttps://github.com/qualixar/superlocalmemory/blob/main/SETUP_FOR_AGENTS.mdstep by step and show me what each step printed. Ask me only one question,before turning on anything optional: whether I also want to find my pictures,screenshots and PDFs by describing them. Connect SuperLocalMemory to the agentI am using. When you finish, run slm doctor and tell me in plain words what works.Alternatively, on macOS and Linux, run one line in Terminal:bashcurl -fsSLo get-slm.sh https://raw.githubusercontent.com/qualixar/superlocalmemory/main/scripts/get-slm.sh && sh get-slm.shFor npm (Node 18+):bashnpm install -g superlocalmemoryslm setupslm connect claude-codeFor pip (Python 3.12–3.14):bashpython3 -m venv .venvsource .venv/bin/activatepython -m pip install superlocalmemoryslm setupslm connect claude-code
License and pricing
Licensed under AGPL-3.0. The project is open source and listed as having no pricing (open source).
Based on the project's public README and repository metadata.