{
  "schemaVersion": "1.0",
  "type": "public-experiment-record",
  "language": "zh-CN",
  "generatedAt": "2026-10-18",
  "publisher": {
    "id": "https://zenleak.cn/#organization",
    "name": "ZenLeak GEO 学院",
    "editorialName": "ZenLeak 编辑部",
    "contact": "geo@allintwo.cloud"
  },
  "interpretationPolicy": {
    "technicalAccessIsNotIndexing": true,
    "indexingIsNotRanking": true,
    "discoveryIsNotCitation": true,
    "missingResultIsNotZero": true
  },
  "record": {
    "id": "machine-entry-consistency-2026-09-17",
    "slug": "machine-entry-consistency-2026-09-17",
    "order": 2,
    "title": "机器入口响应检查：4 类 User-Agent × 4 个公开路径",
    "description": "记录 ZenLeak 对普通研究客户端、Googlebot、bingbot 与 ChatGPT-User 标识请求的 HTTP 响应结果，并解释伪装 User-Agent 测试能够和不能够证明什么。",
    "status": "completed",
    "dateStarted": "2026-09-17",
    "dateCompleted": "2026-09-17",
    "datePublished": "2026-09-17",
    "dateModified": "2026-10-18",
    "nextReviewAt": "2026-10-25",
    "lastRetestedAt": "2026-10-18",
    "lastRetestArtifact": "/downloads/research/machine-entry-retest-2026-10-18.json",
    "executionStatus": "completed",
    "question": "ZenLeak 的首页与三个机器入口是否会因请求中声明的 User-Agent 不同而出现 HTTP 可达性差异？",
    "hypothesis": "如果服务器没有对这些 User-Agent 设置差异化拦截，四个路径应在四类请求标识下都返回 HTTP 200；该结果仍不能证明真实爬虫身份、抓取行为或引用结果。",
    "scope": {
      "sampleSize": 16,
      "unit": "HTTP 请求",
      "target": "4 个 User-Agent 标识与 4 个公开路径的笛卡尔组合",
      "dimensions": [
        "HTTP 状态",
        "请求标识",
        "公开路径",
        "差异化阻断"
      ]
    },
    "methods": [
      "选择首页、robots.txt、llms.txt 和 evidence-index.json 四个公开路径。",
      "分别使用 ZenLeak-Research、Googlebot、bingbot 和 ChatGPT-User 四个 User-Agent 字符串发送独立请求。",
      "记录每个组合的最终 HTTP 状态；本轮只判定可达性，没有把 User-Agent 字符串当作真实爬虫身份。",
      "将 16 个响应按路径与请求标识交叉核对，检查是否存在显式状态码差异。"
    ],
    "questions": [],
    "metrics": [
      {
        "id": "http-200",
        "label": "HTTP 200 响应",
        "value": 16,
        "total": 16,
        "display": "16 / 16",
        "definition": "请求最终返回 HTTP 200；本指标不评价正文是否被真实爬虫使用。"
      },
      {
        "id": "paths-reachable",
        "label": "各标识下均可达的路径",
        "value": 4,
        "total": 4,
        "display": "4 / 4",
        "definition": "每个路径在四种 User-Agent 字符串下都至少获得一次 HTTP 200。"
      },
      {
        "id": "status-blocks",
        "label": "状态码阻断",
        "value": 0,
        "total": null,
        "display": "0",
        "definition": "没有观察到因本轮 User-Agent 字符串变化而产生的非 200 状态。"
      }
    ],
    "observations": [
      {
        "label": "响应状态",
        "finding": "16 个请求全部返回 HTTP 200，未观察到针对四种请求标识的状态码级阻断。",
        "evidenceLevel": "measured"
      },
      {
        "label": "身份边界",
        "finding": "测试客户端只修改了 User-Agent 字符串，没有进行反向 DNS 或平台网络验证，因此不是来自 Google、Bing 或 OpenAI 的真实爬虫请求。",
        "evidenceLevel": "context"
      },
      {
        "label": "抓取与引用",
        "finding": "HTTP 200 只证明测试时能够取回响应，不能证明真实平台已经抓取、索引、训练、采用或引用这些内容。",
        "evidenceLevel": "context"
      },
      {
        "label": "2026-10-01 交叉请求复测",
        "finding": "对首页、robots.txt、llms.txt 和 evidence-index.json 使用普通研究客户端、Googlebot 标签、bingbot 标签和 ChatGPT-User 标签分别请求，共 16 个请求全部返回 HTTP 200，未观察到状态码级阻断。",
        "evidenceLevel": "measured"
      },
      {
        "label": "身份核验边界",
        "finding": "本轮仍只是第一方客户端主动声明不同 User-Agent；没有把这些请求写成真实 Google、Bing 或 OpenAI 爬虫访问。真实身份需要结合服务器日志、官方 IP 范围或正反向 DNS 核验。",
        "evidenceLevel": "context"
      },
      {
        "label": "2026-10-09 交叉请求复测",
        "finding": "对首页、robots.txt、llms.txt 和 evidence-index.json 使用普通研究客户端、Googlebot 标签、bingbot 标签和 ChatGPT-User 标签分别请求，共 16 个请求全部返回 HTTP 200，未观察到状态码级阻断。",
        "evidenceLevel": "measured"
      },
      {
        "label": "2026-10-18 交叉请求复测",
        "finding": "对同一组 4 个公开路径使用 4 个声明的 User-Agent 标签分别请求，共 16 个请求全部返回 HTTP 200，未观察到状态码级阻断。",
        "evidenceLevel": "measured"
      },
      {
        "label": "2026-10-18 身份核验边界",
        "finding": "本轮只更换第一方客户端的 User-Agent 字符串，没有对请求源进行反向 DNS 或官方 IP 核验；不能把结果描述为真实搜索引擎或 AI 爬虫访问。",
        "evidenceLevel": "context"
      }
    ],
    "artifacts": [
      {
        "label": "10 月 1 日机器入口交叉请求复测数据",
        "href": "/downloads/research/machine-entry-retest-2026-10-01.json"
      },
      {
        "label": "10 月 9 日机器入口交叉请求复测数据",
        "href": "/downloads/research/machine-entry-retest-2026-10-09.json"
      },
      {
        "label": "10 月 18 日机器入口交叉请求复测数据",
        "href": "/downloads/research/machine-entry-retest-2026-10-18.json"
      }
    ],
    "limitations": [
      "User-Agent 可以被任意客户端声明，不能用它验证请求真实来源。",
      "本轮只记录 HTTP 状态，没有逐字比较响应正文、缓存层或地区节点差异。",
      "测试覆盖四个代表性路径，不代表站点全部 URL 都不存在访问策略差异。",
      "结果不涉及平台抓取频率、索引资格、AI 训练许可或答案引用。"
    ],
    "conclusions": [
      "所测四个公开路径没有因四种 User-Agent 字符串而出现 HTTP 状态码差异。",
      "2026-10-18 复测仍得到 16/16 个 HTTP 200；这支持入口可达性判断，不支持真实爬虫身份、收录或引用判断。",
      "真实爬虫验证仍应结合官方 IP 验证方法、服务器日志和平台工具，答案引用需要单独的问题面板实验。"
    ],
    "relatedTutorials": [
      "robots-sitemap-llms",
      "ai-crawler-governance",
      "crawl-render-index"
    ],
    "tags": [
      "User-Agent",
      "机器入口",
      "HTTP",
      "爬虫治理"
    ],
    "statusLabel": "已完成",
    "representations": {
      "pagePath": "/experiments/records/machine-entry-consistency-2026-09-17/",
      "pageUrl": "https://zenleak.cn/experiments/records/machine-entry-consistency-2026-09-17/",
      "jsonPath": "/experiments/records/machine-entry-consistency-2026-09-17.json",
      "jsonUrl": "https://zenleak.cn/experiments/records/machine-entry-consistency-2026-09-17.json",
      "markdownPath": "/experiments/records/machine-entry-consistency-2026-09-17.md",
      "markdownUrl": "https://zenleak.cn/experiments/records/machine-entry-consistency-2026-09-17.md"
    }
  }
}