
Search external sources for reusable skills before designing a new one
Provide systematic prompt optimization methodology to improve instruction quality while preserving original characteristics
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
Search external skill sources and evaluate candidates for reuse before creating custom skills
Provide in-depth interview methodology for extracting clear actionable requirements from vague descriptions
Capture repeated A-Team design patterns as reusable instincts for later sessions
Entry point that spawns the Team Architect coordinator to run the full team design workflow
Rewrite generated prompts to be sharper, shorter, and more executable without changing meaning
Check whether a proposed role split is too coarse, too fine, or balanced
Decompose workflow responsibilities into a clean, non-overlapping role set
Decompose workflow responsibilities into a clean, non-overlapping role set
Evaluate whether a proposed agent graph supports safe handoffs and useful parallel work
Provide quantitative methods for evaluating whether agent decomposition granularity is appropriate
Provide methodology and decision framework for decomposing team responsibilities into agent roles with appropriate granularity
Provide framework for analyzing team structure rationality and evaluating collaboration topology efficiency between agents
Instinct-based learning system that captures team design patterns via hooks and evolves them into reusable knowledge
Capture repeated A-Team design patterns as reusable instincts for later sessions
Check whether a proposed role split is too coarse, too fine, or balanced
Apply consistent frontmatter, structure, and path references across authored markdown files
Define universal writing standards, format specifications, and quality requirements for all .md files
Apply consistent frontmatter, structure, and path references across authored markdown files
Rewrite generated prompts to be sharper, shorter, and more executable without changing meaning
Reference library of Anthropic-official prompt engineering patterns, templates, and Claude 4.6 best practices for generation agents
Validate that a generated Codex team is structurally complete and internally consistent
Provide validation methods to check if final team structure output is complete and consistent
Validate that a generated Codex team is structurally complete and internally consistent
Create or improve a skill by using the system skill when available and the legacy local bundle as fallback
Create or improve a skill by using the system skill when available and the legacy local bundle as fallback
Search external sources for reusable skills before designing a new one
Extract precise team requirements through staged, single-direction questioning
Extract precise team requirements through staged, single-direction questioning
Evaluate whether a proposed agent graph supports safe handoffs and useful parallel work