What problem does it solve? PM workspaces run different delivery methodologies (continuous, sprint, cycle, phased) and roadmap groupings, each requiring different frontmatter fields on specs, epics, and initiatives. Without preset detection, authoring agents hardcode assumptions (like always prompting for story points) that produce broken artifacts in workspaces using a different methodology. ## Core Features & Use Cases - Preset Resolution: Reads pm.presets from hero.json and returns the active roadmap, delivery, and overlay combination, degrading to sane defaults (horizon + continuous) when config is missing. - Preset-Conditional Field Rules: Maps each delivery preset to its required fields — points/sprint for sprint, hill_position/cycle for cycle, release/phase for phased — so agents prompt only for relevant fields. - Field Preservation on Switch: Preserves inactive preset fields when a workspace changes methodology, keeping history intact and making switching a low-cost config edit. - Use Case: A story-writer agent in a Shape Up (cycle) workspace loads this skill first, learns the active preset, and prompts for hill_position and appetite instead of story points. ## Quick Start Ask the agent to read the pm.presets configuration from hero.json and tell you which fields to populate for a new feature spec.
