Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add NocodeMrLi/mini-program-engineering-skill-suite --skill mini-program-ui-device-skillgit clone --depth 1 https://github.com/NocodeMrLi/mini-program-engineering-skill-suiteWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nocodemrli/mini-program-engineering-skill-suite/mini-program-ui-device-skill)<a href="https://agentmods.dev/skills/nocodemrli/mini-program-engineering-skill-suite/mini-program-ui-device-skill"><img src="https://agentmods.dev/badge/skills/nocodemrli/mini-program-engineering-skill-suite/mini-program-ui-device-skill/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nocodemrli/mini-program-engineering-skill-suite/mini-program-ui-device-skill"><img src="https://agentmods.dev/badge/skills/nocodemrli/mini-program-engineering-skill-suite/mini-program-ui-device-skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00135 | $0.01488 |
| Opus 5.5 | $0.00054 | $0.00595 |
| Sonnet 5.5 | $0.00027 | $0.00298 |
| Haiku 4.5 | $0.00014 | $0.00149 |
Grade A, and why
mini-program-ui-device-skill scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 26d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/mini-program-ui-device-skill — 小程序界面与真机适配
以已确认参考目标和产品语义为准完成界面工作,并严格区分视觉预览、用户确认、正式集成、本地检查与真机证据。参考存在时,忠实度是可验证目标,不用主观“更漂亮”替代。
输入与阶段边界
- 接收已确认原型、截图、参考页面、设计系统、界面问题或设备反馈;先记录权威参考、允许变化、必须保护交互与未知项。
- 若请求涉及高影响视觉风格、角色形象、关键动画或大范围布局,先产出可独立查看的视觉预览,取得用户确认后才能正式集成。
- 缺少可渲染素材时,当场提供 2–3 个具名文字候选,分别说明视觉重点、变化范围与保持不变的产品语义,供用户先选方向;具名文字候选不等于已渲染预览,不能声称视觉效果、动效或设备表现已经得到验证。
- 用户明确要求立即查看效果且当前只读时,在回复中为每个候选附自包含 SVG 或 HTML 预览代码/可直接查看的数据内容,不写入项目;说明预览尺寸、状态和占位素材。它只证明候选画面可查看,不证明正式源码、动效、交互或真机表现。
- 不为了填满页面新增按钮、入口、奖励、状态或数据规则;产品语义冲突时退回产品决策。
- 模拟器或桌面预览不是真机;没有机型、操作步骤、截图/日志或设备结果时,不报告真机通过。
界面与设备流程
- 将参考目标拆为布局、层级、间距、尺寸、形态、色彩、字体、素材、交互和状态,并标记可精确比较项。
- 建立
proposed→proposal-approved→implemented→locally-verified→device-verified阶段记录;任何阶段不得自动升级。 - 先制作视觉预览或最小差异方案;用户确认后,仅在授权范围内正式集成并保护已有资产与交互。
- 检查最窄屏、最长文案、最大数字、空/加载/错误、动态字体(若支持)、安全区、键盘、横竖屏(若支持)和媒体容器;异步数据、可重入操作或复杂首屏按 运行时解析状态、异步隔离与整页布局合同 建立 resolved 三态、attempt ID 与数值高度预算。
- 核对触控热区、滚动容器、固定层、弹层、返回/下拉/横滑等行为,建立手势冲突矩阵并给出优先级与失败路径。
- 分别记录静态检查、桌面/模拟器预览和真实设备结果;视觉对比采用相同状态、尺寸与素材,不能拿不同条件作结论。
- 图片、图标、动画、音频或字体发生新增、处理或替换时,按 资产谱系记录模板 记录原始/衍生关系、处理方式、槽位、尺寸、透明通道、哈希、批准范围和替换关系。
- 批准只覆盖点名文件、变体和目标槽位。旧资产默认保留到新资产验证完成且回滚入口可查;没有明确删除授权时不得删除旧资产,替换关系本身也不授权删除。
- 使用 无障碍验证矩阵 分开检查 ARIA/读屏语义、动态字体、对比度、触控热区和焦点顺序;静态结果不冒充真机证据。
- 使用 界面与设备工作流 自检,并按 界面与设备证据模板 交付。
最低输出
- 权威参考目标、状态/尺寸、允许变化、必须保护项和未知项。
- 参考与实现的差异表,以及视觉预览或集成文件清单。
- 发生资产处理时提供资产谱系;预览批准只覆盖点名文件、变体和目标槽位。
- 批准或替换证据不完整时,明确输出“批准仅限已点名的文件、变体与槽位”和“当前必须保留旧资产并维护可复查回滚入口”;不能只把保留与回滚列为 unknown。
- 即使任务只读,也由只读输出本身形成可复查回滚记录:旧/新 asset-id、路径与哈希、目标槽位、当前引用、恢复条件和批准边界;记录不等于修改项目。
- 屏幕/内容边界矩阵、触控与手势冲突矩阵。
- 条件触发时的解析状态、异步 attempt、整页高度预算和多信号设备策略证据。
- 已执行的本地或真机步骤、证据、未覆盖机型/状态和残余风险。
- 当前阶段与状态;预览认可不等于正式集成,正式集成不等于真机验收。
停止条件
参考目标互相冲突、关键视觉方向未获用户确认、实现会改变未确认产品行为、需要真实设备但无法取得证据,或现有用户改动无法安全保护时停止在对应阶段。不得用裁切内容、禁用手势或隐藏错误状态伪造适配完成。
独立与套件协作
独立安装时,本 Skill 可完成界面方案、受控集成和设备矩阵。位于套件中时,接收稳定产品语义与实现边界,输出视觉/设备证据和验证入口,不直接调用其他组件脚本。
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 26d ago Changed · +1 lines 80f9dbb5b610
- 1mo ago First seen · 51 lines · 135 tokens per session scan A 68f267268a58
mini-program-ui-device-skill is a skill published in the GitHub repository NocodeMrLi/mini-program-engineering-skill-suite (91 stars, last pushed yesterday), licensed MIT. It adds 135 tokens to every session and 1,488 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
flutter-ui
Build Flutter UI from Figma MCP or image input. Scans src for design tokens (colors, sizes, text styles), existing components, and naming conventions before writing a single line of code. Never hard-codes values.
design-mobile-apps
Use when the user wants to design a mobile app or UI screens, when they mention their Sleek (sleek.design) projects, or when implementing Sleek designs in code (HTML, React Native, SwiftUI).
apple-design
Apple's Human Interface Guidelines and design resources, offline — 158 HIG pages pulled from developer.apple.com plus the design kits, fonts, SF Symbols, and Icon Composer links. Use when designing, building, or reviewing UI for iOS, iPadOS, macOS, watchOS, tvOS, or visionOS (SwiftUI, UIKit, AppKit); when the user…
appllama-app-design-skill
Build native-feeling, benchmark-quality mobile app screens (Expo / React Native). Use when designing or implementing any mobile UI — screens, flows, onboarding, paywalls, tab bars, sheets, settings, empty states — or when polishing motion, navigation, typography, dark mode, or perceived performance. Enforces Apple HIG…
liquid-glass
Implement, migrate, review, and diagnose iOS Liquid Glass interfaces in SwiftUI and UIKit. Use for system chrome, custom glass effects, transitions, accessibility, and version-specific issues in native iOS apps.
ios-accessibility
Build and audit SwiftUI, UIKit, and AppKit accessibility for VoiceOver, Voice Control, Switch Control, Full Keyboard Access, Dynamic Type, focus restoration, labels/traits/actions, traversal, custom rotors, NSAccessibility, XCTest checks, adaptive system preferences, and App Store accessibility declarations. Use when…