设置 Webhook - Pubrio Documentation

Webhook 是接收 monitor 结果的推荐方式。当 monitor 触发时，Pubrio 会实时向你配置的 URL 发送一个包含 JSON 负载的 POST 请求。

前提条件

拥有 monitor 访问权限的 Pubrio API 密钥
一个可公开访问的 HTTPS 端点（或来自 usewebhook.com 的测试 URL）

快速测试： 使用 usewebhook.com 生成免费的临时 webhook URL。你可以在不部署任何东西的情况下检查每个传入负载。

步骤 1：创建带有 Webhook 投递目标的 Monitor

curl -X POST https://api.pubrio.com/monitors/create \
  -H "Content-Type: application/json" \
  -H "pubrio-api-key: YOUR_API_KEY" \
  -d '{
    "name": "My First Monitor",
    "detection_mode": "signal_first",
    "signal_types": ["jobs"],
    "signal_filters": [
      {
        "signal_type": "jobs",
        "filters": {
          "locations": ["US"]
        }
      }
    ],
    "destination_type": "webhook",
    "destination_config": {
      "webhook_url": "https://usewebhook.com/YOUR_WEBHOOK_ID",
      "headers": {
        "X-Custom-Auth": "your-secret-token"
      },
      "body": {
        "pipeline": "my-webhook"
      }
    },
    "max_records_per_trigger": 5,
    "profile_id": 1
  }'

import requests

response = requests.post(
    "https://api.pubrio.com/monitors/create",
    headers={
        "Content-Type": "application/json",
        "pubrio-api-key": "YOUR_API_KEY"
    },
    json={
        "name": "My First Monitor",
        "detection_mode": "signal_first",
        "signal_types": ["jobs"],
        "signal_filters": [
            {
                "signal_type": "jobs",
                "filters": {
                    "locations": ["US"]
                }
            }
        ],
        "destination_type": "webhook",
        "destination_config": {
            "webhook_url": "https://usewebhook.com/YOUR_WEBHOOK_ID",
            "headers": {
                "X-Custom-Auth": "your-secret-token"
            },
            "body": {
                "pipeline": "my-webhook"
            }
        },
        "max_records_per_trigger": 5,
        "profile_id": 1
    }
)

print(response.json())

const response = await fetch("https://api.pubrio.com/monitors/create", {
  method: "POST",
  headers: {
    "Content-Type": "application/json",
    "pubrio-api-key": "YOUR_API_KEY"
  },
  body: JSON.stringify({
    name: "My First Monitor",
    detection_mode: "signal_first",
    signal_types: ["jobs"],
    signal_filters: [
      {
        signal_type: "jobs",
        filters: {
          locations: ["US"]
        }
      }
    ],
    destination_type: "webhook",
    destination_config: {
      webhook_url: "https://usewebhook.com/YOUR_WEBHOOK_ID",
      headers: {
        "X-Custom-Auth": "your-secret-token"
      },
      body: {
        pipeline: "my-webhook"
      }
    },
    max_records_per_trigger: 5,
    profile_id: 1
  })
});

console.log(await response.json());

响应：

{
  "data": {
    "monitor_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "name": "My First Monitor",
    "detection_mode": "signal_first",
    "destination_type": "webhook",
    "is_active": true,
    "is_paused": false,
    "masked_signature": "7••••••••••••••••8df",
    "created_at": "2026-04-06T10:00:00.000Z",
    "signature": "71a2b3c4-d5e6-f789-0abc-def123456789"
  }
}

headers 对象会在每次投递时添加自定义 HTTP 请求头（用于身份验证）。body 对象会在 webhook 负载的根级别添加自定义字段。

保存响应中的 signature — 你需要它来验证传入的负载。它仅在创建时、通过签名揭示端点、或在监控查询中设置 is_signature_reveal: true 时返回。

步骤 2：验证你的 Webhook 连接

使用验证 Webhook 端点测试你的端点是否可达。这会发送一个包含占位数据的示例负载——不消耗积分，不获取真实信号。

curl -X POST https://api.pubrio.com/monitors/webhook/validate \
  -H "Content-Type: application/json" \
  -H "pubrio-api-key: YOUR_API_KEY" \
  -d '{
    "monitor_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "detection_mode": "signal_first",
    "signal_types": ["jobs"],
    "signal_filters": [
      {
        "signal_type": "jobs",
        "filters": { "locations": ["US"] }
      }
    ],
    "destination_type": "webhook",
    "destination_config": {
      "webhook_url": "https://usewebhook.com/YOUR_WEBHOOK_ID",
      "headers": { "X-Custom-Auth": "your-secret-token" },
      "body": { "pipeline": "my-webhook" }
    },
    "profile_id": 1
  }'

成功的响应会返回发送的示例请求负载以及你的端点返回的响应——这样你可以在上线前确认连接正常。

步骤 3：使用真实数据测试

连接验证通过后，使用处理尝试端点触发一次真实运行。这会获取实际信号并投递到你的 webhook——使用 tried_at 设置一个近期的过去时间以确保有可用数据：

curl -X POST https://api.pubrio.com/monitors/process/try \
  -H "Content-Type: application/json" \
  -H "pubrio-api-key: YOUR_API_KEY" \
  -d '{
    "monitor_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "tried_at": "2026-01-01T00:00:00.000Z",
    "profile_id": 1
  }'

与 validate 不同，try 端点运行真实扫描并消耗积分。使用它来验证真实负载正确到达，并在计划扫描启动前快速预估结果。

步骤 4：验证签名

每个 monitor 都有唯一的签名，用于验证传入的负载确实来自 Pubrio。

curl -X POST https://api.pubrio.com/monitors/signature/reveal \
  -H "Content-Type: application/json" \
  -H "pubrio-api-key: YOUR_API_KEY" \
  -d '{
    "monitor_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "profile_id": 1
  }'

{
  "data": {
    "monitor_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "signature": "71a2b3c4-d5e6-f789-0abc-def123456789"
  }
}

将此签名与传入负载中的 monitor.monitor_id 进行比较以验证真实性。

Webhook 负载结构

负载根据 monitor 的 detection_mode 而不同：

信号优先
公司优先

在 signal_first 模式中，负载包含顶层的 signals 数组：

{
  "monitor": {
    "monitor_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "name": "My Signal Monitor",
    "detection_mode": "signal_first",
    "signal_types": ["jobs", "news"],
    "signal_filters": [...],
    "company_filters": {...},
    "is_company_enrichment": true,
    "is_people_enrichment": true,
    "people_enrichment_configs": [...]
  },
  "metadata": {
    "total_signals": 3,
    "total_companies": 2,
    "total_people": 5
  },
  "triggered_at": "2026-04-05T20:29:43.832Z",
  "signals": [
    {
      "signal_type": "jobs",
      "signal": {
        "signal_type": "jobs",
        "job_search_id": "...",
        "companies": [
          {
            "domain_search_id": "...",
            "company_name": "...",
            "domain": "...",
            ...
          }
        ],
        ...
      },
      "companies": [
        {
          "domain_search_id": "...",
          "company_name": "...",
          "domain": "...",
          "logo_url": "...",
          "country_code": "...",
          "company_size": 5000,
          "industry": "...",
          "people": [...],
          "emails": [...],
          "phones": [...],
          ...
        }
      ]
    },
    {
      "signal_type": "news",
      "signal": {
        "signal_type": "news",
        "news_search_id": "...",
        "news_id": "...",
        "title": "...",
        "summary": "...",
        "published_at": "...",
        "source": "...",
        "category": "...",
        "companies": [...],
        ...
      },
      "companies": [...],
      ...
    },
    ...
  ],
  "pipeline": "my-webhook"
}

每个信号条目包含信号详情以及关联的已补充公司和人员。

在 company_first 模式中，负载包含顶层的 companies 数组，其中含有已补充的公司数据和嵌套的信号：

{
  "monitor": {
    "monitor_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "name": "My Company Monitor",
    "detection_mode": "company_first",
    "signal_types": ["jobs", "news", "advertisements"],
    "signal_filters": [...],
    "company_filters": {...},
    "is_company_enrichment": true,
    "is_people_enrichment": true,
    "people_enrichment_configs": [...]
  },
  "metadata": {
    "total_signals": 4,
    "total_companies": 2,
    "total_people": 8
  },
  "triggered_at": "2026-04-03T17:45:27.228Z",
  "companies": [
    {
      "company_name": "Acme Corp",
      "domain": "acmecorp.com",
      "domain_search_id": "...",
      "country_code": "US",
      "logo_url": "...",
      "linkedin_name": "acmecorp",
      "company_size": 5200,
      "industry": "Enterprise Software",
      "estimated_revenue": 50000000,
      "founded_year": 2010,
      "company_address": "San Francisco, CA",
      "specialties": ["SaaS", "Cloud Computing", ...],
      "linkedin_url": "https://linkedin.com/company/...",
      "locations": ["US"],
      "signals": [
        {
          "signal_type": "news",
          "signal": {
            "news_id": "...",
            "title": "Acme Corp Launches New AI Product",
            "summary": "...",
            "published_at": "2026-04-03T16:35:00.000Z",
            "source": "techcrunch.com",
            "category": "launches",
            "news_category_name": "Product Launch",
            ...
          }
        },
        {
          "signal_type": "jobs",
          "signal": {
            "job_search_id": "...",
            ...
          }
        },
        ...
      ],
      "people": [
        {
          "name": "Jane Smith",
          "title": "VP of Engineering",
          "email": "[email protected]",
          ...
        },
        ...
      ],
      "emails": ["[email protected]", ...],
      "phones": ["+14155551234", ...],
      "contacts": [...],
      ...
    },
    ...
  ],
  "pipeline": "my-webhook"
}

数组中的每家公司包含完整的补充档案、所有匹配的信号和补充的人员联系方式。

来自 destination_config 的自定义 body 字段会出现在根级别（例如上面示例中的 "pipeline": "my-webhook"）。

扩张信号

除 jobs、news 和 advertisements 之外，监控还可以侦测扩张信号——即表明某公司正在进入或深耕新市场的带日期证据。只需将 expansions 加入 signal_types：

{
  "name": "进入美国市场",
  "detection_mode": "signal_first",
  "signal_types": ["expansions"],
  "signal_filters": [
    {
      "signal_type": "expansions",
      "filters": {
        "tos": ["US"],
        "stages": ["expanding", "scaling"],
        "signal_strengths": ["high", "very_high"],
        "window_days": 30
      }
    }
  ],
  "destination_type": "webhook",
  "destination_config": {
    "webhook_url": "https://usewebhook.com/YOUR_WEBHOOK_ID"
  }
}

扩张筛选条件使用 Expansion Search 的词汇表，而非职位/新闻/广告那一套——用 froms 与 tos 指定走廊，另有 stages、momentum、freshness、scopes、signal_types、signal_strengths 和 window_days。有效标识请通过 Expansion Reference 获取。

扩张信号按公司与市场分组，而非按单条信号分组。一家公司同时进入两个市场时会产生两条记录，每条各自携带该市场的信号时间线。

扩张信号负载

{
  "monitor": {
    "monitor_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "name": "进入美国市场",
    "detection_mode": "signal_first",
    "signal_types": ["expansions"],
    "signal_filters": [...]
  },
  "metadata": {
    "total_signals": 2,
    "total_companies": 1,
    "total_people": 0
  },
  "triggered_at": "2026-08-14T09:12:04.551Z",
  "signals": [
    {
      "signal_type": "expansions",
      "signal": {
        "signal_type": "expansions",
        "expansion_signal_id": 90211,
        "domain_search_id": "550e8400-e29b-41d4-a716-446655440002",
        "country_code": "US",
        "signal_type_slug": "HIRE",
        "signal_type_name": "Hiring",
        "signal_subtype_slug": "country_manager",
        "signal_strength_slug": "high",
        "polarity": "expansion",
        "event_date": "2026-08-13T14:00:00.000Z",
        "event_date_precision": "day",
        "source_type": "linkedin",
        "display_label": "Hired Country Manager",
        "evidence_url": "https://linkedin.com/company/example-corp",
        "metadata": {...},
        "companies": [
          {
            "domain_search_id": "550e8400-e29b-41d4-a716-446655440002",
            "company_name": "Example Corp",
            "domain": "example.com",
            "stage_name": "Expanding",
            "stage_description": "...",
            "home_country": "GB",
            "home_country_name": "United Kingdom",
            "target_country": "US",
            "target_country_name": "United States",
            "estimated_employees": 420,
            "linkedin_url": "https://linkedin.com/company/example-corp"
          }
        ]
      },
      "companies": [...]
    }
  ]
}

各字段的含义详见扩张字段参考。若希望按需拉取相同的数据行而非依赖触发，请使用 Expansion Signal Search。

电子邮件投递目标

对于偏好电子邮件投递的团队，将 destination_type 设置为 "email"：

{
  "destination_type": "email",
  "destination_config": {
    "email": "[email protected]"
  }
}

Pubrio 为代理商和团队提供白标邮件投递支持。联系我们了解自定义发件人域名和品牌。

故障排除

Webhook 未接收到负载

确认你的端点是可公开访问的（不在防火墙或 VPN 后面）
确保它返回 200 状态码——其他状态码会被视为失败
使用验证 Webhook 端点测试连通性
检查统计日志中的错误信息和响应码

Monitor 在失败后暂停

如果你的 webhook 持续返回非 200 状态码，monitor 在达到 max_failure_trigger 次连续失败后会暂停。修复问题后通过更新 Monitor 重新激活。

重复负载

如果投递失败且配置了重试，你可能会多次收到相同的负载。使用 triggered_at 或日志 ID 在你端进行去重。

负载过大

减小 max_records_per_trigger 以限制每次投递的记录数。你也可以缩小筛选范围以减少匹配的信号量。

​前提条件

​步骤 1：创建带有 Webhook 投递目标的 Monitor

​步骤 2：验证你的 Webhook 连接

​步骤 3：使用真实数据测试

​步骤 4：验证签名

​Webhook 负载结构

​扩张信号

​扩张信号负载

​电子邮件投递目标

​故障排除

前提条件

步骤 1：创建带有 Webhook 投递目标的 Monitor

步骤 2：验证你的 Webhook 连接

步骤 3：使用真实数据测试

步骤 4：验证签名

Webhook 负载结构

扩张信号

扩张信号负载

电子邮件投递目标

故障排除