Menu
Anti-Fake Data & Anti-Regex Reality Enforcement Layer — 阻止 AI 用 MOCK/regex/静态数据伪造系统可运行性，强制 AI 在不确定时停下来问而非编造假系统
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
对任意技术对象（GitHub 项目/论文/技术文章/产品方案/License/数据集）执行六维批判式评审：拆解主张、攻击实验与方法漏洞、证据校验、专家评分，输出 Quick/Expert/Academic 报告。触发时机：用户要求点评/批判/审计/毒舌评价/评审/打分任何技术对象，或说「这个项目/论文/方案靠不靠谱」「值不值得学/投/用」「帮我审一下」「毒舌点评一下」，或调用 hos-critic-review。
HOS 自动化工作流 — 基于 Dify 平台的 AI 报告撰写自动化工具
Structured prompt engineering framework with JSON templates for generating standardized, reusable AI prompts across diverse scenarios including articles, exams, projects, security, and operations.
人机验证通行技能 — 安全测试中的人机识别与验证机制研究
方法论驱动的 AI 原生安全测试引擎
AI 原生产品工程技能系统 — 四技能集成，最小化 Token 成本、上下文爆炸和维护负担
| name | HOS-Silly-Mock |
| description | Anti-Fake Data & Anti-Regex Reality Enforcement Layer — 阻止 AI 用 MOCK/regex/静态数据伪造系统可运行性，强制 AI 在不确定时停下来问而非编造假系统 |
| version | 1.0.0 |
| author | HOS-Sec-Engine |
| license | MIT |
| compatibility | ["claude-code","cursor","windsurf","github-copilot","trae-cn"] |
| metadata | {"tags":["code-integrity","mock-detection","regex-abuse","reality-binding","silent-failure","ai-code-quality","enforcement","code-review","static-analysis","quality-gate"],"category":"code-review","subCategory":"code-integrity-enforcement","risk-level":"high","confidence":0.94} |
HOS-Silly-Mock 不是一个"优化代码"的 Skill，而是一个 Reality Enforcement Layer（真实强制执行层）。 它阻止 AI 使用 MOCK 数据、正则表达式、静态数据来伪造系统可运行性，强制 AI 在不确定时"停下来问"，而不是"编一个能跑的假系统"。
❌ 阻止 AI 用"编造确定性"替代"不确定真实系统"
✅ 强制 AI 诚实表达不确定性
| # | 威胁 | 表现 | 本质 |
|---|---|---|---|
| 1 | MOCK污染 (Mock Leakage) | API没接→写假数据, UI没后端→填demo json, 表单没验证→写固定值 | AI 用"编造确定性"替代"不确定真实系统" |
| 2 | Regex硬编码依赖 (Regex Overuse Bias) | JSON parse→regex, HTML→regex, log→regex, URL→regex | 用"局部规则"替代"系统语义" |
| 3 | 沉默失败 (Silent Failure) | 不报错、继续写完整系统、但系统其实是假的 | AI 主动制造"不可维护的假系统" |
| 4 | 数据来源单薄 (Reality Binding Gap) | 变量无来源、transform无输入、sink无context | 逻辑链断裂导致无意义代码 |
mock data, fake data, hardcoded data, dummy data, test data in production, regex parse,
regex abuse, silent failure, fake system, demo mode, mock leakage, 假数据, 硬编码,
正则滥用, 沉默失败, 假系统, 静态数据, reality binding, data authenticity,
source-transform-sink, MOCK_MODE, 缺少 API, mock污染
mock, fake, dummy, sample, demo, testDataTODO: connect to real APIreality-check, fake-data-detector, mock-leakage-guard, regex-reflection-blocker,
silent-failure-killer, anti-fake-code, code-integrity-enforcer, 真实连接强制,
反假数据层
┌──────────────────────────────────────────────────────────┐
│ HOS-Silly-Mock │
│ │
│ Layer 1: MOCK显性化强制器 (Mock Exposure Layer) │
│ 任何 mock 必须显式标注 ⚠ MOCK_MODE: TRUE │
│ │
│ Layer 2: Regex禁用反射器 (Regex Reflection Blocker) │
│ 禁止 regex 用于结构化数据解析，强制使用标准解析器 │
│ │
│ Layer 3: 真实连接强制器 (Reality Binding Layer) │
│ 每个变量必须绑定 source → transform → sink 三元组 │
│ │
│ Layer 4: 沉默失败检测器 (Silent Failure Detector) │
│ 检测完整但无真实意义的逻辑链 │
└──────────────────────────────────────────────────────────┘
MOCK_MODE: TRUE 显式标注/.../）和 new RegExp(match, replace, split, exec 的调用上下文throw / reject 的错误路径JSON.parse/<tag[^>]*>(.*?)<\/tag>/gmock 关键词（变量名可能是 mockService 但合理）Array.from({length: 100}, (_, i) => ({id: i}))）// @silly-mock:allow-test 豁免// ❌ 错误: 无标注的 mock 数据
const users = [
{ id: 1, name: 'Alice', email: 'alice@example.com' },
{ id: 2, name: 'Bob', email: 'bob@example.com' },
{ id: 3, name: 'Charlie', email: 'charlie@example.com' },
];
// ✅ 正确: 显式标注 + 来源说明
/**
* MOCK_MODE: TRUE
* reason: /api/users not deployed yet, will replace in phase 2
* @silly-mock:allow
*/
const users = await fetch('/api/users').catch(() => {
console.warn('[MOCK] /api/users unavailable, using fallback data');
return [
{ id: 1, name: 'Alice', email: 'alice@example.com' },
];
});
// ❌ 错误: 使用 regex 解析 JSON
const jsonStr = '{"name":"Alice","age":30}';
const nameRegex = /"name"\s*:\s*"([^"]+)"/;
const match = jsonStr.match(nameRegex);
const name = match?.[1];
// ✅ 正确: 使用 JSON.parse + schema validation
import { z } from 'zod';
const UserSchema = z.object({ name: z.string(), age: z.number() });
const jsonStr = '{"name":"Alice","age":30}';
const data = UserSchema.parse(JSON.parse(jsonStr));
const name = data.name;
// ❌ 错误: 使用 regex 解析 HTML
const html = '<div class="user">Alice</div>';
const nameRegex = /<div[^>]*class="user"[^>]*>([^<]*)<\/div>/;
const match = html.match(nameRegex);
const name = match?.[1];
// ✅ 正确: 使用 DOMParser
const parser = new DOMParser();
const doc = parser.parseFromString(html, 'text/html');
const name = doc.querySelector('.user')?.textContent;
// ❌ 错误: 无 source 的变量
const users = transformData(someData);
// users 的来源 someData 未定义
// ✅ 正确: 完整 source → transform → sink
const rawUsers = await api.getUsers(); // source
const processed = rawUsers.map(normalizeUser); // transform
renderUserList(processed); // sink
// ❌ 错误: 沉默失败 — 完整但不真实的系统
function processOrders() {
const orders = getOrders(); // 可能失败但不处理
const total = calculateTotal(orders);
const updated = updateStock(orders);
return { total, updated };
// ❌ 没有 try/catch, 没有 throw, 没有 error handling
}
// ✅ 正确: 错误显式化
async function processOrders() {
try {
const orders = await getOrders();
if (!orders.length) throw new Error('No orders to process');
const total = calculateTotal(orders);
const updated = await updateStock(orders);
return { total, updated };
} catch (err) {
console.error('Order processing failed:', err);
throw err; // 或明确的 fallback
}
}
╔═══════════════════════════════════════════════╗
║ HOS-Silly-Mock Reality Report ║
╠═══════════════════════════════════════════════╣
║ Reality Score: ████████░░ 82/100 ║
║ Data Authenticity: ███████░░░ 75/100 ║
║ ═══════════════════════════════════════════ ║
║ Mock Leakage Risk: LOW ║
║ Regex Abuse Risk: LOW ║
║ Silent Failure Risk: NO ║
║ Reality Binding: PASS ║
╠═══════════════════════════════════════════════╣
║ Findings: 3 warnings, 0 errors ║
║ ├─ [L1-MOCK] users may contain mock data ║
║ │ → add MOCK_MODE annotation ║
║ ├─ [L2-REGEX] 1 regex may parse structured ║
║ │ → replace with DOMParser ║
║ └─ [L4-SILENT] 1 function lacks error path ║
║ → add try/catch or throw ║
╚═══════════════════════════════════════════════╝
MOCK_MODE: TRUEgetStaticProps 构建时数据被标记MOCK_MODE: TRUE + reason@silly-mock:allow-testYou are HOS-Silly-Mock Enforcement Engine.
Your role is to prevent fake system construction.
Rules:
1. NEVER use mock data unless explicitly marked MOCK_MODE: TRUE
2. NEVER use regex for structured parsing
3. NEVER continue logic without real I/O source
4. NEVER silently simulate system completeness
If any of the following occurs:
- missing API / data source
- fake data used without declaration
- regex used for structured parsing
→ STOP generation immediately
→ Request real input or system boundary clarification
You must prefer:
- HALT over hallucination
- INCOMPLETE over fake completeness
- ERROR over silent fake success