pmo-skill-refinerlisted

# PMO Skill Refiner A narrower refinement wrapper over an Anthropic scaffolding skill that injects PMO-platform awareness and runs the preserved eval/optimization harness. ## When to use vs. skip **Use this skill when:** - Creating a new skill that will live under `release/skills/<name>/` (a PMO-platform skill, not a generic skill). - Iterating on an existing PMO skill's eval set, trigger description, or benchmark — specifically the rigor work (quantitative eval, variance analysis, blind A/B, description-trigger optimization). - Injecting missing PMO-platform fields into a skill that pre-dates the 7-field injection discipline. **Skip and route elsewhere when:** - Generic non-PMO skill creation → route to `anthropic-skills:skill-creator` directly. - Modifying existing skills' SKILL.md structure absent eval work (adding a mode, reorganizing sections, editing Output Contract shape) → route to `pmo-skill-editor` directly. - Creating a plugin with commands, hooks, MCP servers (not just a skill) → route to `cowork-plugin-management:create-cowork-plugin` directly. The refiner's unique value-add is the PMO injection layer plus the preserved eval rigor (variance analysis, cross-skill false-positive detection, description-trigger optimization loop) on PMO-suite skills. Anywhere else is over-applying the tool. ## Use When Common operator phrasings that route to this skill (preserved as trigger-matching examples for the description-trigger optimization loop): **For new-skill crea