Generate MDC-based Claude rules and skills from failure patterns.
npx skills add https://github.com/udecode/dotai --skill writing-skills-udecode
Please help me install this Agent Skill. Skill: writing-skills Source: https://github.com/udecode/dotai/tree/main/.claude-plugin/plugins/skills/skills/writing-skills Command: npx skills add https://github.com/udecode/dotai --skill writing-skills-udecode
This meta-Skill applies Test-Driven Development (TDD) principles to the creation and refinement of AI Skills, ensuring they are robust, clear, and resistant to rationalization. It prevents the deployment of ineffective skills by systematically identifying failure patterns and closing loopholes.
Use the writing-skills skill to create a new skill for managing project dependencies.
High-intent search queries and answers about installing and using this skill.
Test-Driven Development for AI skills means identifying failure patterns first (RED), writing the skill to address them (GREEN), then refactoring to close loopholes. This approach ensures skills are robust and resistant to misinterpretation, making them more effective guides for other AI agents.
Generate MDC-based skills by defining frontmatter fields (name, description, alwaysApply), writing rules in .claude/rules/, generating context-loaded skills in .claude/skills/, and validating changes via npx skiller@latest apply. This process produces skills that integrate directly into your Claude workflow.
Ineffective skills often fail because they allow rationalization—users find loopholes to ignore guidance. This skill bulletproofs against that by systematically documenting failure patterns upfront, then writing explicit rules that close common excuses, ensuring skills actually guide behavior.
Claude Search Optimization (CSO) guidelines help craft skill names, descriptions, and keywords to maximize discoverability by other AI agents. Clear naming and strategic keyword selection ensure your skills surface when agents search for relevant capabilities.
Yes. This skill supports updating existing rules in .claude/rules/ (both alwaysApply: true and false) and merging skills into AGENTS.md. Use TDD principles to identify where existing rules fail, then refactor and validate via npx skiller.
The skill works with .claude/rules/ for rule files, .claude/skills/ for context-loaded skills, and AGENTS.md for merged documentation. It supports glob patterns for file matching and enforces consistent frontmatter fields across all generated content.
Wide-MoatRun RED-GREEN-REFACTOR compliance tests on agent Skills under pressure scenarios.
KanuToCLTest AI agent skills using Red-Green-Refactor and pressure simulations.
MRWillisTValidate SKILL.md entries using a RED-GREEN-REFACTOR workflow against pressure scenarios.
ivanrvpereiraApply test-driven RED-GREEN-REFACTOR cycles to author and verify Skills.
