Menu
Use to score generated prompts, blueprints, code, products and multi-agent systems.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Use when output risks looking like a generic app-builder app, landing, dashboard, or SaaS.
Use when converting a raw brief or spec into a Vertical Blueprint Contract for an AI app builder or coding agent.
Use when an AI builder receives this repository as guidance for turning incomplete requests into specific products.
Use when building iOS, Android, Expo, Flutter, SwiftUI, Jetpack Compose, or mobile-first apps.
Use when designing multi-agent orchestration, control towers, agentic operations, or tool-calling systems.
Use when selecting the correct project profile and blueprint before writing a specification.
| name | specification-driven-coding-evaluation-scorecard |
| description | Use to score generated prompts, blueprints, code, products and multi-agent systems. |
Use docs/18-evaluation-scorecards.md and the relevant file in scorecards/.
Never mark an output as ready without stating the remaining gaps.
Based on SOC occupation classification
