lead-scraper

自动抓取潜在客户信息 - 智能线索抓取工具，支持多平台客户信息采集、清洗、去重、分类

Security Scan

The skill's documentation describes multi‑platform web scraping (including sites that normally require credentials or anti-bot handling) but provides no code, no install, and no guidance about required credentials or how scraping is performed — this mismatch and under-specification merit caution.

Purpose & Capability

The name/description (lead scraping across many platforms) matches the SKILL.md goals. However it claims support for platforms (e.g., LinkedIn, Tianyancha, 企查查) that typically require authentication, scraping headers/cookies, or legal/ToS handling. The skill declares no required credentials or tools, which is underspecified: either the instructions assume the agent already has the ability to access those platforms, or necessary credentials/details are omitted.

Instruction Scope

SKILL.md is instruction-only and provides example natural-language commands for scraping, cleaning, scoring and exporting. It does not tell the agent to read arbitrary local system files except example data-cleaning of a named file (contact_list.xlsx), and it states compliance constraints (robots.txt, no CAPTCHA bypass). But the instructions are high-level and vague about implementation (how to fetch pages, handle rate limits, CAPTCHAs, authenticated sessions, or where data is transmitted), which gives the agent wide discretion and could lead to broad web requests or data collection behaviors.

Install Mechanism

No install spec and no code files — lowest install risk. Nothing will be written to disk by the skill itself during install because there is no installer.

Credentials

The skill requests no environment variables or credentials, yet claims integration with services that commonly require API keys, session cookies, or logins. That omission is an inconsistency: a realistic multi‑platform scraper typically needs credentials or at least cookies/proxies and rate-limit controls. Additionally, the SKILL.md includes a crypto payment address (USDC on Polygon) and WeChat pay info — unrelated to operation but useful to know; monetization info is not a technical requirement but should be considered when trusting the author.

Persistence & Privilege

Flags show always:false and no special config paths or persistent privileges. The skill does not request permanent presence or modify other skills/configs.

What to consider before installing

This skill is an instruction-only description of a scraper rather than an implementation; that makes it low install-risk but leaves important gaps. Before installing or enabling: (1) verify the developer/company identity and reputation, and whether there is real code backing these claims; (2) do NOT provide platform credentials (LinkedIn, Tianyancha, etc.) unless you understand how they will be stored/used; (3) treat the compliance statements as promises only — confirm how robots.txt, rate limits, and CAPTCHA handling are enforced; (4) if you must test, run it in a controlled environment and restrict the agent's ability to exfiltrate data or access sensitive files; (5) consider declining to enable autonomous invocation or limit the skill to manual invocation until you obtain an actual implementation or source code; (6) be aware of legal/privacy risks when scraping personal or commercial contact data and consult legal/compliance if needed.

线索抓取

智能潜在客户信息采集工具，帮助企业快速建立销售线索库。

功能特性

• 🎯 多平台抓取: 企查查、天眼查、B2B平台、社交平台
• 🔧 数据清洗: 格式标准化、无效数据过滤、重复去重
• 📊 智能分类: 按行业、地区、规模、意向度分类
• 📤 多格式导出: Excel、CSV、JSON、CRM对接
• ⭐ 线索评分: 自动评估线索质量和优先级
• 🔐 合规安全: 只抓取公开信息，遵守法律法规

使用方式

# 按行业抓取
抓取 杭州地区 电子商务行业 企业联系方式

# 按关键词抓取
抓取 包含"智能制造"关键词的企业信息

# 按平台抓取
从LinkedIn抓取 软件工程师 线索

# 展会线索
抓取 2026年广交会 参展商信息

# 数据清洗
清洗这份客户名单 去重补全 contact_list.xlsx

# 线索评分
评估这份线索质量 leads.csv

配置要求

必需配置

无特殊配置要求。

可选配置

{
  "skills": {
    "entries": {
      "lead-scraper": {
        "config": {
          "platforms": ["qichacha", "tianyancha"],
          "rate_limit": 100,
          "output_format": "xlsx"
        }
      }
    }
  }
}

支持平台

企业信息平台

• 企查查
• 天眼查
• 启信宝
• 国家企业信用信息公示系统

B2B平台

• 阿里巴巴1688
• 慧聪网
• 中国制造网

社交平台

• LinkedIn
• 脉脉

行业资源

• 展会参展商名录
• 行业协会会员
• 招投标信息平台

输出示例

========================================
           线索抓取报告
========================================

【抓取概况】
  关键词: 电子商务 杭州
  数据来源: 企查查、天眼查
  抓取数量: 256条
  有效线索: 243条 (94.9%)

【线索统计】
  按行业分布:
    - 电子商务: 45%
    - 软件服务: 25%
    - 物流仓储: 15%
    - 其他: 15%

  按企业规模:
    - 大型企业: 5%
    - 中型企业: 25%
    - 小型企业: 70%

  按地区分布:
    - 西湖区: 30%
    - 滨江区: 25%
    - 余杭区: 20%
    - 其他: 25%

【优质线索 Top 10】
  1. 杭州XX科技有限公司
     评分: 9.5/10 | 行业: 电商服务 | 规模: 中型
     联系人: 张总 | 电话: 138****1234
     意向度: 高 | 备注: 近期有采购需求

  2. 浙江YY电子商务有限公司
     评分: 9.2/10 | 行业: 跨境电商 | 规模: 中型
     联系人: 李经理 | 电话: 139****5678
     意向度: 高 | 备注: 扩张期

  ...

【导出文件】
  格式: Excel (.xlsx)
  路径: ./output/leads_20260402.xlsx
  包含字段: 公司名称、联系人、电话、邮箱、地址、行业、规模、评分

【跟进建议】
  优先联系评分>8.0的线索
  最佳联系时间: 工作日 10:00-11:30, 14:00-17:00
  推荐话术: [已生成话术模板]

========================================

合规声明

✅ 允许

• 抓取公开信息
• 遵守robots.txt
• 尊重隐私设置

❌ 禁止

• 抓取非公开信息
• 破解验证码
• 恶意高频请求
• 倒卖个人信息

注意事项

1. 抓取前请确认目标平台的使用条款
2. 数据质量受源头影响
3. 遵守《个人信息保护法》和《数据安全法》
4. 获得数据主体同意后再营销

更新日志

v1.0.0 (2026-04-01)

• 初始版本发布
• 支持多平台抓取
• 实现智能清洗去重
• 支持多格式导出

---

开发者: 盈指量杭州科技有限公司
定价: ¥99/次 或 ¥299/月订阅
收款: USDC (Polygon: 0x3DbFf9E97b10a10d4A2079B4273473da7e6F4120) | 微信支付