Agent Memory

Gives your Claude Code agent persistent memory across conversations. The repository doesn't include documentation, but the premise is straightforward: it stores context, decisions, and learned preferences so you're not constantly re-explaining your project structure or coding conventions. With over 1,100 installs since February, it's clearly solving a real pain point. The lack of a SKILL.md is frustrating if you want to understand exactly what gets remembered and how to configure it, but for anyone tired of repetitive context-setting at the start of each session, it's worth trying. Think of it as giving Claude a notebook instead of amnesia.

Install to Claude Code

npx -y skills add api/git --skill agent-memory --agent claude-code