What problem does it solve? Choosing a technology stack for a greenfield project is slow and error-prone when done ad hoc. This Skill turns a written PRD into a concrete starter recommendation and a machine-readable hand-off file that a downstream bootstrapper can consume to scaffold the project. ## Core Features & Use Cases - PRD-driven stack selection: Reads context/foundation/prd.md frontmatter priors (product type, scale, timeline) and feature signals from functional requirements to constrain the decision. - Curated starter registry with quality gates: Reasons over language-aware starter cards in references/starter-registry.yaml, applying four agent-friendly filters (typed, convention-based, popular in training data, well-documented) across JS, Python, Ruby, Java, Go, Rust, PHP, .NET, and Dart. - Two interview paths: A standard path that short-circuits to a vetted recommended default, and a custom path with a full residual interview, framework-variant comparison, and a five-point self-check. - Use Case: After writing a PRD for a solo SaaS MVP, invoke the skill to get a recommended starter (e.g., Astro + Supabase + Cloudflare), confirm deployment and CI/CD choices, and produce context/foundation/tech-stack.md ready for the bootstrapper. ## Quick Start Ask the assistant to pick a tech stack for the project based on the PRD at context/foundation/prd.md.