Use when executing implementation plans with independent tasks in the current session
Formal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles
Research-before-coding workflow. Search for existing tools, libraries, and patterns before writing custom code.
Use when context is growing large, responses are slow, or you need to optimize token budget across system prompt, background processes, and model selection
A comprehensive verification system for Claude Code sessions.
Use when implementing any feature or bugfix, before writing implementation code
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Suggests manual context compaction at logical workflow breakpoints to preserve context through task phases rather than losing it to arbitrary auto-compaction