Persistent memory for Claude that survives across conversations. It writes knowledge to `.claude/skills/agent-memory/memories/` so you can save research findings, architectural decisions, codebase gotchas, and in-progress work that would otherwise disappear when the context resets. The proactive usage guidance is spot on: save things that took effort to discover, check memories before diving into related work. It's straightforward infrastructure for anyone who's tired of re-explaining the same context or rediscovering the same solutions. With 468 installs and passing all three security audits, it's clearly solving a real pain point for developers working on longer projects.
npx skills add https://github.com/yamadashy/repomix --skill agent-memory