ceorkm/kratos-cli ? reverse-engineered prompt

Reverse engineered prompt

Build me a CLI tool called kratos-memory that gives AI coding agents persistent local memory between sessions.

I want to be able to run it with npx in any project, save useful notes, search them later, and get a compact context summary the agent can inject at the start of a session. It should keep everything local in SQLite, support full text search, and have separate memory storage per project so different repos do not mix. Please include commands for saving, searching, asking natural language questions, recent items, getting a memory by id, deleting, status, switching projects, scanning for secrets or personal data, and installing hooks for Claude Code and Codex so memory can be captured automatically.

Make it feel polished and safe, with JSON output for automation, helpful defaults, and no cloud dependency or API keys. If you need to check current docs for Claude Code or Codex hook setup, look them up online.

Are you gonna build this?

make sure you review the code using coderabbit

Try freeSponsored — opens CodeRabbit in a new tab