SeanJ1ang GitHub avatar

design-judge-skills

SeanJ1ang

Design Judge Skills provides modular, evidence-driven AI agent skills covering the full design award workflow—from case retrieval and evaluation to award matching, entry preparation, and submission check.

Stars

503

7-day growth

+239

Forks

14

Open issues

0

License

Apache-2.0

Last updated

2026-07-28

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It uniquely combines official-source-first design principles, transparent scoring, and broad coverage of major design awards (iF, Red Dot, IDEA, DIA, etc.) with over 22,000 observed award entries as contextual reference data.

Who it is for

  • Design award applicants (companies, studios)
  • Individual designers and students preparing submissions
  • Design educators and advisors guiding students or clients
  • Developers building custom award workflow agents

Use cases

  • Plan a complete award application pipeline with handoff records
  • Retrieve and verify winning cases in the same category from official sources
  • Evaluate a design project (student concept or professional) with transparent scoring and critical risk flags
  • Compare award fit across multiple programs (iF, Red Dot, IDEA, DIA, etc.) and generate submission texts

Strengths

  • Always prioritizes official award pages over third-party summaries for rules and cases
  • Clearly separates user facts, model inferences, and items needing confirmation
  • All scores and fit assessments are transparently derived, not claiming to predict winning probability
  • Modular design allows independent use of each skill (search, evaluate, match, prepare, check)

Considerations

  • Several skills are in Beta (noted boundary issues); stable only for design-award-search
  • The 22,125 observed records provide descriptive background only, cannot be used to estimate winning chances
  • Does not simulate unpublished jury preferences or internal award processes, and requires runtime verification of deadlines/fees/eligibility from official sites

README quick start

Design Judge Skills

面向设计奖全流程的证据驱动 Agent Skills:从获奖案例检索、设计评价和奖项匹配,到申报文字准备与提交终检。

快速开始 · 获奖样本体量 · 安装 · 技能索引 · 贡献与开发 · Star 历史 · English

design-judge-skills 将设计奖申报拆解为边界清晰、可独立触发与验证的技能模块,围绕官方来源、证据定位与透明评分构建可追溯的辅助决策流程,并依据公开标准比较申报路径、解释适配度与确定申报优先级。

目录

1. 项目介绍

Design Judge Skills 覆盖设计奖申报中的五类核心任务:

  • 检索并核验同类别获奖案例;
  • 用证据化量表评价设计本体与展示表达;
  • 匹配奖项、项目路径和申报类别;
  • 从用户材料中准备有来源约束的申报文字;
  • 按当届官方规则检查提交包是否就绪。

当前配置覆盖 iF DESIGN AWARD、iF DESIGN STUDENT AWARD、Red Dot Product Design、Red Dot Design Concept、IDEA、DIA、K-Design、GOOD DESIGN AWARD Japan、Core77、James Dyson 和 EPDA。涉及截止日期、费用、资格、类别与提交规格时,技能要求在运行时重新核验官方页面。

评价模块内置来自 iF、iF Student、Red Dot 与 IDEA 的 22,125 条获奖或入围作品聚合观察记录。这些数据只提供描述性背景,不改变核心评分,也不能用于估计获奖概率。参见覆盖范围、隐私与限制

2. 工作流

设计材料 ── 不确定路线或需要完整流程 ──> design-award-pipeline
  │
  ├─ 寻找同类获奖案例 ──────────────> design-award-search
  ├─ 评价设计与展示表达 ────────────> design-evaluation
  └─ 选择奖项 / 赛道 / 类别 ───────> design-award-match
                                         │
                                         v
                                design-information-prep
                                         │
                                         v
                                design-submission-check

这些模块可以独立使用。完整申报流程通常从评价或匹配开始,在目标奖项确定后进入信息准备,最后执行提交终检。

3. 设计原则与边界

  1. 优先使用一手来源。 奖项规则和案例以官方页面为准,搜索摘要与第三方页面只能用于发现线索。
  2. 事实与推断分离。 用户材料中的事实、模型推断和待用户确认项必须分别标记。
  3. 评分保持透明。 适配度和评价分数用于辅助决策,不代表获奖概率。
  4. 模块职责单一。 检索、评价、奖项匹配、文字准备和提交检查不互相越界。
  5. 不冒充官方判断。 可对齐公开评审标准,但不模拟未公开的评委偏好或内部流程。

4. 快速开始

安装完成后,直接把设计图片、展板、说明书、项目 brief、研究材料或申报文件交给 Agent,并说明任务。下面的提示词可以直接复制。

想做什么直接这样说
规划完整参奖流程使用 $design-award-pipeline,根据现有材料规划完整参奖路线,并维护各阶段的交接记录。
查找同类获奖案例使用 $design-award-search,为这个康复训练产品寻找经过官方页面核验的同类别获奖案例。
评价学生概念`使用 $design-evaluation 评价附件中的设计。成熟度由我确定为“学生概念”,请分别输出设计本体、展示

Description

Evidence-driven Agent Skills for design award research, evaluation, award matching, entry writing, and submission readiness.

Related repositories

Similar projects matched by category, topics, and programming language.

7-e1even
Featured
7-e1even GitHub avatar

learn-agent

A collection of notes on building coding agents, derived from the production development of the desktop agent Reina, with each mechanism simplified into a zero-dependency, single-file Node.js demo.

AI & Machine LearningLarge Language Models
218
makecindy
Featured
makecindy GitHub avatar

cindy

Cindy is an open-source AI agent that runs locally on your machine, integrates multiple AI harnesses and models, and provides memory, skills, and automation to perform real work in your projects and apps.

AI & Machine LearningLarge Language Models
958
0xwilliamortiz
Featured
0xwilliamortiz GitHub avatar

ponytail-improved

Ponytail is a plugin for AI coding agents that enforces a disciplined ladder of reuse before writing code, reducing code volume by roughly 54% while preserving safety.

AI & Machine LearningAI Agents
545