What problem does it solve? Writing an AI agent skill that triggers correctly, stays portable across harnesses, and actually improves behavior is hard to do by intuition alone. This Skill provides a disciplined workflow for creating, revising, migrating, and auditing skills so they meet the Agent Skills specification and are backed by real evidence rather than guesswork. ## Core Features & Use Cases - End-to-end skill authoring workflow: Eight guided steps from resolving intent through drafting, structural validation, behavioral testing, description tuning, and packaging. - Independent review and audit rubric: A detailed review checklist covering triggering, instruction economy, evidence integrity, and portability, plus a mechanical signal-scan script for first-pass checks. - Validation templates: Ready-to-use templates for baseline behavior comparisons, trigger query testing, and new skill scaffolding. - Use Case: You want to add a new skill to your team's shared catalog. Use this Skill to draft the SKILL.md from a template, validate it against the Agent Skills specification, run matched behavioral comparisons, test the description against trigger queries, and produce an installable package. ## Quick Start Use the creating-portable-skills skill to audit my existing skill package and produce a prioritized list of fixes.
