Create, refine, or debug SKILL.md modules using a test-driven approach.
npx skills add https://github.com/kaissalha/webld --skill skill-writing-kaissalha
Please help me install this Agent Skill. Skill: skill-writing Source: https://github.com/kaissalha/webld/tree/main/.agents/skills/skill-writing Command: npx skills add https://github.com/kaissalha/webld --skill skill-writing-kaissalha
Skill Writing provides a structured, test-driven approach to creating, refactoring, and validating agent instruction modules (SKILL.md) so agents follow desired behavior with documented guardrails and rationalizations.
Document and implement a test-driven skill for agent prompt formatting, starting from a failing scenario and iterating until the agent complies.
High-intent search queries and answers about installing and using this skill.
Prevent prompt drift by creating SKILL.md modules using a test-driven approach, writing a failing scenario first, then implementing and refactoring the agent instructions until compliance is achieved.
Your agent might ignore instructions if the SKILL.md module lacks explicit guardrails and rationalizations; debugging and optimizing the prompt structure with a test-driven approach resolves this non-compliance.
Test-driven development for AI agent skills is a structured approach where you write a failing scenario first, then implement and refactor the SKILL.md module until the agent reliably follows the desired behavior with documented guardrails.
Structure a SKILL.md file by including frontmatter with a name and description, a concise operational description, and guardrails with explicit failure handling to ensure reliable prompt loading and activation across projects.
The best way to refactor existing agent instructions is to apply a test-driven approach: write failing scenarios for current behavior, then refine the SKILL.md prompt structure, descriptions, and rationale to improve compliance and test coverage.
Yes, you need guardrails with explicit failure handling for every agent instruction module to ensure reliable prompt handling, reduce misbehavior, and maintain consistent activation across projects.
sigridjinethGenerate hierarchical AGENTS.md documentation across a codebase.
robit-manRuns mutation testing to evaluate how well existing test suites detect introduced defects.
billydeeii136Compute a weighted 0-10 codebase health score and track trends over time.
zhbonerDraft SKILL.md files and refine them through prompt-based test evaluations.
