seanj1ang/design-judge-skills

Analysis updated 2026-05-18

★ 228PythonAudience · designerComplexity · 2/5LicenseSetup · easy

Mindmap

mindmap
  root((repo))
    What it does
      Searches past winners
      Evaluates designs
      Matches award categories
      Drafts entry text
      Checks submission readiness
    Tech stack
      Python
      Claude Code skills
      Codex skills
    Use cases
      Design award research
      Entry writing help
      Submission compliance checks
    Audience
      Designers
      Design studios
    Setup
      Install via npx skills
      Works in Claude Code or Codex

mindmap root((repo)) What it does Searches past winners Evaluates designs Matches award categories Drafts entry text Checks submission readiness Tech stack Python Claude Code skills Codex skills Use cases Design award research Entry writing help Submission compliance checks Audience Designers Design studios Setup Install via npx skills Works in Claude Code or Codex

What do people build with it?

USE CASE 1

Research verified past winners in the same design category before entering an award.

USE CASE 2

Get a structured evaluation of a design and its presentation materials before submitting.

USE CASE 3

Compare several design awards to see which categories best fit a specific project.

USE CASE 4

Draft award entry text from project materials while tracking which facts are confirmed.

What is it built with?

PythonNode.jsClaude CodeCodex

How does it compare?

	seanj1ang/design-judge-skills	django-haystack/celery-haystack	hellogithub-team/hydra
Stars	228	227	231
Language	Python	Python	Python
Last pushed	—	2026-07-06	—
Maintenance	—	Active	—
Setup difficulty	easy	moderate	moderate
Complexity	2/5	3/5	2/5
Audience	designer	developer	developer

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · easy Time to first run · 5min

Requires Node.js 18 or higher to install skills via the npx skills command.

Licensed under Apache 2.0, which allows free use, modification, and commercial use as long as you keep the copyright notice.

In plain English

This repository is a set of AI agent skills built to help designers and studios prepare entries for international design awards, such as iF Design Award, Red Dot, IDEA, and several others. It is meant to plug into agent tools like Claude Code or Codex, giving the AI a structured way to help with the whole award submission process instead of answering one off questions. The skills are broken into separate modules, each handling one piece of the process. One module searches for and verifies similar past winning entries from official award sources. Another evaluates a design and its presentation materials against a scoring rubric, separating facts from the AI's own inferences. A third compares different award tracks and categories to suggest which ones fit a project best. A fourth helps draft the written text required for a submission, again keeping user supplied facts separate from anything the AI infers. A final module checks a finished submission package against that year's official rules and gives a go, conditional go, or no go verdict. A pipeline skill can chain these together automatically when someone is not sure which one to use. The project states clearly that it favors official award sources over search summaries or third party pages, and that its scoring is meant to support a person's decision, not predict whether an entry will actually win. It also includes a dataset of over 22000 observed past entries used only as background context, not as a way to guess odds of winning. Installation happens through a command line tool called npx skills, which can install the skills globally or per project into supported agent tools including Claude Code, Codex, Cursor, and several others. Once installed, a user simply describes their task in plain language, such as asking it to compare which awards fit a product, and the right skill is triggered. The project is aimed at designers, design studios, and teams managing design award submissions, and assumes basic comfort installing command line tools and using an AI coding agent.

Copy-paste prompts

Prompt 1

Use design-award-search to find verified past winners in my product category for iF Design Award.

Prompt 2

Use design-evaluation to critique the attached design concept and flag any critical issues.

Prompt 3

Use design-award-match to compare Red Dot, IDEA, and Core77 fit for this project.

Prompt 4

Use design-information-prep to draft IDEA entry text from these project files and list missing facts.

Prompt 5

Use design-submission-check to review this submission package and give a go or no-go verdict.

Frequently asked questions

What is design-judge-skills?

A collection of AI agent skills that help designers research, evaluate, and prepare entries for design awards like iF, Red Dot, and IDEA.

What language is design-judge-skills written in?

Mainly Python. The stack also includes Python, Node.js, Claude Code.

What license does design-judge-skills use?

Licensed under Apache 2.0, which allows free use, modification, and commercial use as long as you keep the copyright notice.

How hard is design-judge-skills to set up?

Setup difficulty is rated easy, with roughly 5min to a first successful run.

Who is design-judge-skills for?

Mainly designer.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.