Agent skill · software engineering · jeremylongshore
mnemos
Persistent memory and learning-loop skills for AI coding agents using mnemos MCP tools. Use whenever mnemos_* MCP tools are available and you are doing work worth remembering. Cross-session memory, conventions, corrections, and architectural decisions surface back into context automatically. Triggers on session start (record a goal), session end (record a summary), the user says "save", "remember", "record this", "we were wrong about", and on any genuine correction or architectural decision. Keeps mnemos's memory store alive across sessions so the next session starts smarter. Without this skill, agents silently edit and the store goes empty.
Why this skill is useful
Adds persistent memory capabilities through mnemos MCP tools, enabling agents to retain and recall important decisions and corrections across sessions.
What it needs
Requires mnemos installed locally. About 7k tokens when loaded. Last updated 2026-08-07. 2,604 stars on the source repository.
What this skill does
mnemos Overview You have a persistent memory layer available via the mnemos MCP tools. By default you will skip these tools on plain editing tasks. Don't. Mnemos is the learning-loop primitive for the project you are working on: corrections, decisions, and conventions compound across sessions if you record them. Silently editing leaves the store empty and the next session blind. There is a known correction already stored about this exact failure mode: "agent skipped mnemossessionstart on editing tasks — LLMs skip optional tool calls when the task looks like plain reading/editing; the agent needs an external nudge." This skill is that nudge. Prerequisites The mnemos binary is on PATH. mnemos doctor reports green for binary, config, storage, and each agent registration. mnemos init has wired the MCP server into your agent host (Claude Code, Claude Desktop, Cursor, Windsurf, or Codex CLI). The mnemos tools are visible this session. If they are not, the MCP server is not connected — see Error Handling. Instructions Session lifecycle Start. The Claude Code SessionStart hook runs mnemos prewarm, which opens a mnemos session by default and injects a mnemossessionid: line into context. If that ID is present, reuse it for sessionid fields and do not call mnemossessionstart again. If no ID is present (manual MCP setup, non-Claude client, or hook disabled) and you are about to do real work, open one: Without a goal, the UserPromptSubmit hook backfills the first real prompt when available. With a goal, the session becomes a durable record that mnemos replay and future prewarms can use. End. When the user signals done ("ship it", "that's it", "commit and close", "we're done"), close the session: Status values: ok failed blocked abandoned. An open session with no summary is dead weight in the next session's prewarm. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills jeremylongshore/mnemos