The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Continuous Learning v2 Spec
This document captures the v2 continuous-learning architecture:
- Hook-based observation capture
- Background observer analysis loop
- Instinct scoring and persistence
- Evolution of instincts into reusable skills/commands
Primary implementation lives in:
skills/continuous-learning-v2/scripts/hooks/
Use this file as the stable reference path for docs and translations.