michaelschecht Design Award Evaluation
Treats "is this design any good" as a sourced research question, not a taste call, by grounding every judgment in verified precedent from official award-program sources.
Prerequisites
- Node.js 18+ and the
npx skillsinstaller (also deployable to Codex, OpenCode, and other agent platforms) - Reference material for the design being evaluated (images, spec sheet, or product description)
When To Use
- Deciding which award programs actually fit a design before spending effort on an application
- Getting an evidence-backed quality/risk assessment of a design rather than a subjective opinion
- Finding verified, officially-sourced precedent cases in the same category to benchmark against
- Preparing or double-checking a submission package against a program's actual rules
Usage
- Route the request through the pipeline skill first when it's unclear which stage is needed — it decides the minimum set of stages to run.
- Search for verified precedent: pull comparable award-winning or shortlisted works, cross-checked against official award pages rather than secondhand write-ups.
- Run the evaluation skill for an evidence-backed score and risk report on the design itself.
- Run the match skill to compare candidate award programs (e.g. iF Student vs. Red Dot Design Concept) and rank submission priority.
- Use the information-prep skill to extract and draft the factual content an application needs.
- Before submitting, run the submission-check skill against the target program's actual rules.
Best Practices
- Treat official award pages as the source of truth; use general search results only as supporting context, never as the final citation.
- Don't skip straight to evaluation — precedent search first gives the evaluation something concrete to benchmark against.
- Re-run the submission check after any late edit to the application material, not just once at the start.
References
- Repository: https://github.com/SeanJ1ang/design-judge-skills