What problem does it solve? Writing documentation-style skills for AI agents often produces untested instructions that agents ignore or rationalize away under pressure, leading to unreliable behavior in production workflows. ## Core Features & Use Cases - TDD-Based Skill Authoring: Applies the RED-GREEN-REFACTOR cycle to skill creation by running baseline pressure scenarios with subagents before writing any documentation. - Rationalization Bulletproofing: Provides rationalization tables, red flag lists, and explicit loophole counters so discipline-enforcing skills resist agent workarounds. - Discovery Optimization (CSO): Defines naming, description, and keyword rules so future agents can find the right skill from its frontmatter alone. - Use Case: When creating a new skill like a code-review checklist, run baseline scenarios without the skill, document how agents fail, write the minimal skill addressing those failures, then re-test until agents comply under time, sunk-cost, and authority pressure. ## Quick Start Ask the agent to help you create a new skill by first running baseline pressure scenarios with subagents, then writing and iteratively testing the SKILL.md until agents comply.
