Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/YuDefine/nuxt-supabase-starternpx agentmods add rules/yudefine/nuxt-supabase-starter/review-gui-surfaceWrote 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/rules/yudefine/nuxt-supabase-starter/review-gui-surface)<a href="https://agentmods.dev/rules/yudefine/nuxt-supabase-starter/review-gui-surface"><img src="https://agentmods.dev/badge/rules/yudefine/nuxt-supabase-starter/review-gui-surface/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/rules/yudefine/nuxt-supabase-starter/review-gui-surface"><img src="https://agentmods.dev/badge/rules/yudefine/nuxt-supabase-starter/review-gui-surface.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 | $0.00000 | $0.11441 |
| Opus 5 | $0.00000 | $0.05720 |
| Sonnet 5 | $0.00000 | $0.02288 |
| Haiku 4.5 | $0.00000 | $0.01144 |
Grade A, and why
review-gui-surface scanned grade A with 1 finding 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 2d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`http://127.0.0.1:5174` 只准 agent 探測（`curl /api/changes`）。人開的入口永遠是上一表，**NEVER** 因「user 在本機」改 host。 How it starts
The opening of the file, as written. The whole thing — 430 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review-GUI Surface SoP
核心命題：vendor/scripts/review-gui.ts 本體經 fixtures gate hardened 後，incident 漂移到外圍 agent surface。本 rule 對所有呼叫 review-gui 的 surface 統一規約。
本 rule 是 [[agent-self-verification]] 的特例化（review-gui 是其中一個 evidence collection surface），同時延伸 [[manual-review]] 對 review-gui 互動的規約。
為什麼這條 rule 存在
2026-05-{22,23} 累積 3 條 surface pitfall，共通失敗模式：surface agent / wrapper 把 review-gui 當「黑箱」，沒對其 contract 做主線預檢與後驗：
- [[pitfall-handoff-mode-b-skips-review-gui-scan]]
- [[pitfall-verified-ui-compound-item-single-screenshot-evidence-gap]]
- [[pitfall-review-gui-detail-page-no-impl-gate]]
適用範圍
任何呼叫 review-gui flow 的 agent / wrapper / handoff branch：
| Surface | 入口 | 預期 contract |
|---|---|---|
/commit 0-MR gate block |
plugins/hub-core/skills/commit/SKILL.md Step 0-MR |
block 後 MUST auto-triage pending items（MUST 9）；active runtime 可處理的先自行推進，只有 bucket=ready 才引導 user 到 review-gui |
/handoff Mode B 2B.0 |
plugins/hub-core/skills/handoff/SKILL.md Step 2B.0/2B.1.7 |
推薦 user 跑 review:ui 前 MUST 先跑 review-gui.ts --scan 寫入 HANDOFF.md |
screenshot-review verify mode |
主線派 reviewed visual-evidence worker（per [[agent-routing]]） | item 含 compound visual state → 分成 scoped sub-items 或 multi-screenshot annotation |
verified-ui evidence collection |
vendor/snippets/verify-channels/ui-final-state-brief*.template.md |
compound state evidence 必拆 / 必標多 screenshot |
screenshot-review subagent 的 CLI 呼叫 |
agent body 內的 runtime-approved browser CLI | invoke 前 verify CLI contract（per [[agent-self-verification]] § MUST 4） |
| review-gui detail page 互動 | vendor/scripts/review-gui.ts server-side handlers |
impl 完成率 < threshold → manual review block readonly + amber banner（已 implemented v1.4.30+） |
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.
- 2d ago Changed · -24 lines 40007c0220eb
- 3d ago Changed · +2 lines baeea0ff2e9e
- 4d ago Changed 9b0c7ebc8daf
- 7d ago First seen · 452 lines · 0 tokens per session scan A 120160ca8676
review-gui-surface is a cursor rule published in the GitHub repository YuDefine/nuxt-supabase-starter (45 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 11,441 tokens. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other cursor rules, from other repositories
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
control-plane-descriptors
Control plane descriptor and instance implementation patterns.