What problem does it solve? Untested skill documentation often fails in practice because agents rationalize around rules under time, sunk-cost, or authority pressure, and authors cannot know whether a skill teaches the right behavior without watching agents fail without it first. ## Core Features & Use Cases - TDD-based skill authoring: Applies the RED-GREEN-REFACTOR cycle to documentation by running baseline pressure scenarios with subagents, writing the skill to address observed failures, then closing loopholes. - Discovery optimization (CSO): Provides rules for writing trigger-focused descriptions, keyword coverage, naming conventions, and token-efficient skill structure. - Bulletproofing guidance: Includes rationalization tables, red-flag lists, and persuasion-principle research to make discipline-enforcing skills resistant to workarounds. - Use Case: Before deploying a new TDD-enforcement skill, run a scenario where an agent has 200 lines of untested code and a dinner deadline, document its excuses verbatim, then write the skill to counter exactly those rationalizations. ## Quick Start Ask the agent to create a new skill for a recurring technique, starting with baseline pressure scenarios before writing any documentation.
