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 mythkiven/figma-ios-codegen --skill figma-ios-screenshot-verificationgit clone --depth 1 https://github.com/mythkiven/figma-ios-codegenWrote 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/mythkiven/figma-ios-codegen/figma-ios-screenshot-verification)<a href="https://agentmods.dev/skills/mythkiven/figma-ios-codegen/figma-ios-screenshot-verification"><img src="https://agentmods.dev/badge/skills/mythkiven/figma-ios-codegen/figma-ios-screenshot-verification/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/mythkiven/figma-ios-codegen/figma-ios-screenshot-verification"><img src="https://agentmods.dev/badge/skills/mythkiven/figma-ios-codegen/figma-ios-screenshot-verification.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00051 | $0.01288 |
| Opus 5 | $0.00026 | $0.00644 |
| Sonnet 5 | $0.00010 | $0.00258 |
| Haiku 4.5 | $0.00005 | $0.00129 |
Grade A, and why
figma-ios-screenshot-verification 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 10d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Figma 视觉还原比对（Phase 4，强制）
代码生成完成后必须执行；用户明确写「无需比对」时方可豁免（须在回复注明）。
⛔ 前置硬约束：先做节点对账，再做视觉比对
视觉比对天然存在盲区：截图能验"画得对不对"，但验不出"少画了什么"——少了一整个区块时，剩下的区块视觉上仍可能"看起来挺对"，造成假阳性的"基本对齐"判断。
因此进入视觉比对之前，必须先完成 figma-ios-commercial-delivery §0「节点覆盖率对账」 并满足以下任一：
- ✅ 节点覆盖率 = 100%
- ✅ 缺失节点全部属于
figma-ios-codegen-workflow§「全量还原原则」「唯一允许的简化场景」三类，且每条已注明依据
如果节点覆盖率不达标：
- 立即停止视觉比对（不要先比已实现部分、不要"边比边补"）
- 按 workflow 的「未完成处置模板」声明缺失节点
- 回到阶段 2 补齐代码
- 补齐后再重新进入本 skill
禁止反模式：
- ❌ 「视觉差异不大，缺失部分可后续补」——视觉差异低 ≠ 节点齐全
- ❌ 「主要区块已对齐，标签组/段位等可继续完善」——这是把缺失伪装成增量
- ❌ 跳过节点对账直接发"视觉还原比对报告"——会让用户以为已经全量交付
数据来源
- 截图：
{data_dir}/screenshot.png（figma-ios-preload-data阶段 1 已下载） - 节点数据：
{data_dir}/design.json（用于核对颜色/字体/尺寸等"事实值"）
不再调用 MCP
get_screenshot或自行截图。
比对维度
| 维度 | 检查要点 | 常见问题 |
|---|---|---|
| 整体布局 | 各区块位置、间距、对齐 | 层级拍平、坐标偏移、漏渲染 |
| 导航栏 | 标题/按钮/高度/背景 | 高度 44 vs 88 弄错 |
| 颜色 | 背景/文字/边框 | 绕过 color_map / 未用 fallback |
| 字体/字重 | 字号、粗细 | Regular/Medium 混淆 |
| 切图 | 位图位置/比例 | imageset 未拷或路径错 |
| iconfont | 图标类型/尺寸/颜色 | 用了错误的 symbol |
| 底部安全区 | 吸底按钮 | 被 Home Indicator 遮挡 |
| 状态切换 | 选中/未选中视觉差异 | 未实现 updateAppearance() |
输出报告（强制写入交付物）
## 视觉还原比对报告
### ✅ 一致项
- 整体布局：与设计稿一致
- 导航栏：高度用 host.utils.status_bar_nav_height，标题居中
- 颜色：背景 `MKUIStyle.mk_c10()`、文字 `MKUIStyle.mk_c2()`（以 color_map 为准）
- iconfont：查 iconfont_map.json，与稿一致
### ⚠️ 差异项（已处置）
- 「步骤标签」字号：稿 16px，代码生成时取 design.json 实测 14px。已按数据包为准。
- 圆角：图层 1:289 在截图上看似 16，design.json corner_radius=16，已对齐。
### ❌ 无法还原项
- 渐变文字（bg-clip-text + transparent）：UIKit 不原生支持，已用 CAGradientLayer + mask 模拟（轻微差异）
- 服务器下发图片：mock 占位，实际由业务接口注入
差异处置规则
| 差异类型 | 处置 |
|---|---|
| 代码 bug（颜色/坐标写错） | 立即修复，重新比对 |
| 数据包字段与截图冲突 | 以数据包为准（已固化优先级），在报告中说明 |
| 技术限制（MeshGradient 等） | 若有切图则用 imageset；无则代码近似 + 报告说明 |
| 动态内容 | 占位 + 注释说明注入点 |
自检清单
- 前置：节点覆盖率对账已通过（100% 或缺失项全部豁免）
- 已读
{data_dir}/screenshot.png - 按 8 个维度逐一比对
- 报告分 ✅ / ⚠️ / ❌ 三块
- 差异已处置或明确告知用户
- 没有调用 MCP get_screenshot
- 报告里没有出现"差异不大但缺了 X，如需要可补"等降级话术（见 commercial-delivery 「回复纪律」）
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.
- 10d ago First seen · 91 lines · 51 tokens per session scan A 6562cd7306af
figma-ios-screenshot-verification is a skill published in the GitHub repository mythkiven/figma-ios-codegen (9 stars, last pushed 2mo ago), licensed MIT. It adds 51 tokens to every session and 1,288 once invoked, about $0.0003 per session on Opus 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 2026-08-31.
Other skills, from other repositories
e2e-testing
AI-powered E2E testing for any app — Flutter, React Native, iOS, Android, Electron, Tauri, KMP, .NET MAUI. Connects via MCP to running apps so the agent can take screenshots, tap elements, enter text, scroll, inspect UI trees, and verify state with natural language. Use when the user wants to test an app's UI…
Appium Mobile Testing
Mobile application testing skill using Appium for iOS and Android, covering device capabilities, selectors, gestures, and cross-platform testing strategies.
Espresso Android Testing
Android UI testing with Espresso framework including view matchers, actions, assertions, idling resources, and RecyclerView testing patterns.
ios-rocketsim-operator
RocketSim iOS Simulator UI: inspect and control accessibility state, gestures, typing, hardware buttons, and CLI automation.
mobile-emulator-test
QA a native Android build end-to-end on the emulator. Use for "test on emulator", "QA Android build", "verify native build", "white screen", "cache rehydration", "Expo dev-client QA", "adb reverse". Pairs Metro/adb walk with Supabase + Sentry MCPs for three-layer CRUD verification.
mobile-testing
When writing E2E tests for mobile or configuring Detox.