capn-hook - Persistent memory for coding agents

Show HN: Capn-hook for coding agents – don't grep the same mystery twice

capn-hook provides persistent memory for coding agents like Claude Code and Codex, preventing them from re-exploring the same codebase mysteries in every session. By caching file discoveries and automatically invalidating them when source code changes, it ensures agents always access up-to-date context without redundant token usage. This local-first tool integrates via simple hooks, allowing developers to chart expensive discoveries once and recall them instantly, saving significant time and computational resources on repeat questions.

"Don't grep the same mystery twice."