mngerasimenko/keepware-llm-lab

keepware's engineering work on LLM agents: agent memory stored in markdown files instead of a vector database; skills and prompts.

1 file for Claude Code: keepware-llm-lab CLAUDE.md — 525 tokens loaded in every session.

Open on GitHub

3Stars on the repository

1Files it configures its agents with

525Tokens loaded in every session

1Agent configured

Instructions

keepware-llm-lab CLAUDE.md CLAUDE.md A 525 tok

These files are mngerasimenko/keepware-llm-lab's own configuration — they tell Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.