What problem does it solve? Agent skills often fail because descriptions don't trigger discovery, rules get rationalized away under pressure, and monolithic files waste context tokens. This Skill provides a decision-tree dispatcher with templates, standards, and testing methodology to create skills that agents actually find and follow. ## Core Features & Use Cases - Tiered Architecture Guidance: Routes you to Tier 1 (single-file), Tier 2 (multi-file with references), or Tier 3 (platform-scale) structures based on complexity. - Discovery Optimization (CSO): Teaches trigger-based descriptions, keyword strategy, and naming conventions so skills get selected from large libraries. - Anti-Rationalization & Testing: Provides RED-GREEN-REFACTOR pressure-testing methodology and persuasion principles to bulletproof discipline-enforcing rules. - Use Case: You wrote a TDD skill but agents keep skipping tests under deadline pressure. Use this Skill to run baseline pressure scenarios, capture rationalizations verbatim, and add explicit counters until compliance holds. ## Quick Start Ask the agent to help you create a new skill for a specific task, and it will walk you through template selection, description optimization, and testing.
