design-direction
Skill2 starsChanged 53 days agoMITView on GitHub
Explore distinct visual directions and refine generic AI-generated interfaces using tool-generated random strings, concrete creative briefs, screenshot critique, and selective media. Use for alternative design concepts or making a bland interface distinctive. Route visual previews to image generation and runnable changes to code; keep small fixes and faithful reproduction scoped.
8 other projects use a version of this.
Adopted by
Most recently taken up by
- Littlpinguin/marketing-cockpit-templatetheir own version, first seen Nov 3, 2026
- w2ur/claude-code-setuptheir own version, first seen Oct 13, 2026
- sukirman1901/frontend-experttheir own version, first seen Sep 10, 2026
- fernando-delrio/business-crewtheir own version, first seen Sep 6, 2026
- zrosenfield/sharepoint-ai-skillstheir own version, first seen Aug 4, 2026
- danielmachado60/claude-setuptheir own version, first seen Aug 1, 2026
Other versions8 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 8
- zrosenfield/sharepoint-ai-skills★ 96same name
- Littlpinguin/marketing-cockpit-template★ 9same name
- sergeeey/Claude-cod-top-2026★ 9same name
- sukirman1901/frontend-expert★ 9same name
- fernando-delrio/business-crew★ 2same name
- jangles-byte/atelier★ 2same name
- w2ur/claude-code-setup★ 2same name
- danielmachado60/claude-setup★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
Files
Design direction
把技巧变成可执行操作。先判断交付方式,再判断设计问题;对实际画面做判断,保留具体选择依据。
1. Agent 入口:判断用户要生图还是代码
每次执行先记录 deliverable_mode,并用一句话告诉用户本次交付什么。此规则是执行 Agent 的必经判断,不能只放在示例中。
| 用户意图 | 模式 | 交付 |
| 看效果、三版视觉、首屏方案、效果图、用生图模型 | image | 独立设计图片;不启动网站实现 |
| 修改现有项目、实现页面、需要能点击、修复交互 | code | 沿用技术栈实现并验证 |
| 先看方案,再落地 | image-then-code | 先图片,方向选定后再实现;已有选定稿则直接进入实现 |
| 只讨论、分析方法、写建议 | advice | 文字分析 |
用户明确选择优先。只说“优化首页”时,从项目、上文、交付动词判断;缺失信息确实改变交付时才问一个问题。不要强制代码任务先做三张图,也不要把视觉探索升级成代码项目。
图片按钮只是视觉设计,不能声称交互已验证。代码中的生成图是素材,正文和控件仍使用真实界面元素。生图不可用时说明缺口,不静默换成代码画图或付费 API。
2. 建立最小基准
确定目标用户、核心任务、当前页面、品牌与功能约束、范围、方案数量和画幅。 改官网或已有产品:读取页面并查看实际首屏截图,记录 URL、日期、视口、状态。将截图作为生图真实参考输入;文本读取不等于视觉检查。受阻时说明缺口,不虚构看过的画面。 只加载相关资料。信息足够就复述假设并执行,不重复访谈。各候选保持核心内容、主要行动和画幅一致,优先改变构图与阅读顺序。
3. 按问题读取执行配方
| 问题 | 配方 |
| 方案趋同或需要多方向 | 随机字符串:实际运行脚本,再解释启发 |
| “高级、漂亮”过于抽象 | 创意简报:感受转成可见决定 |
| 需要发现画面差距 | 独立评审:隔离上下文,有限迭代 |
| 缺少视觉素材或需要完整效果图 | 图片 |
| 复杂运动能解释产品 | 视频:循环叠层或连续转场 |
| 拥挤、模板感、文案空泛 | 收尾:删减、产品关联、文案核对 |
用户指定技巧时实际执行并保留证据。随机字符串只用于启发,不推翻需求。不机械跑完八项。生图使用当前可用的 imagegen 技能和工具;代码沿用项目;截图使用被允许的浏览器。不同时加载多套互相冲突的总流程。
4. 检查与停止
默认先检查一轮,最多两轮针对性修正;用户预算优先。停止依据是目标、重大问题和预算,不追求主观满分。 独立代理仅在当前指令允许或用户授权时调用;否则标明“自检”,不假称独立评审。图片模式只检查视觉内容;代码模式实际操作主要交互和目标尺寸。
5. 交付与记录
图片:展示全部独立方案,保留稳定名称、提示词、种子、参考图和最终路径。注明概念效果图;用户选择后再处理实施请求。 代码:提供实现位置、验证和限制。不要擅自发布或改线上站点。 工作区用简短记录保存模式理由、参考、种子、方向、提示词、评审与路径;小任务一个文件即可。项目临时偏好不写回 Skill。 出处与新增规则见 来源边界。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

