What problem does it solve? Writing feature specifications that are architecturally sound, properly scoped, and ready for implementation is hard; this Skill enforces a staff-engineer-grade process with a hard Open Questions gate so specs never proceed on unexamined assumptions. ## Core Features & Use Cases - Skeleton-first drafting: Start with a TLDR plus key sections and a numbered Open Questions block that acts as a hard stop before research or design begins. - Structured spec output: Produces a standardized spec covering problem statement, architecture, data model, API contracts, edge cases, risks, and a phased implementation plan where every step leaves the app working. - Severity-ranked architectural review: Audits existing specs with findings ranked Critical/High/Medium/Low against the repository's own agent instructions and canonical primitives. - Autonomous mode: An --autonomous flag resolves gate questions with the most reversible, lowest-blast-radius defaults for unattended pipeline runs. - Use Case: A developer needs a spec for a new caching layer; the Skill drafts a skeleton, blocks on critical unknowns like cache invalidation strategy, then produces a phased plan that maps directly onto an automated PR-creation workflow. ## Quick Start Ask the agent to write a spec for your feature idea, providing a short brief describing what you want to build.
