← 返回 Skills 市场

江苏盐城盐南、经开17个网站招标信息

Name: 江苏盐城盐南、经开17个网站招标信息
Author: xzc13815555922

作者 xzc13815555922 · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ✓ 安全检测通过

总下载

当前安装

版本数

在 OpenClaw 中安装

/install bidding-assistant

功能描述

自动采集盐南高新区、经开区招投标信息；当用户需要定时采集招标网站、筛选特定区域项目、生成PDF日报/月报、推送报到飞书时使用

使用说明 (SKILL.md)

招投标信息采集技能

概述

自动化采集盐南高新区、经开区相关招投标信息，支持：多网站并行采集、智能区域筛选、数据去重存储、PDF日报/月报生成、飞书推送。

触发场景：用户需要监控招标信息、采集特定区域项目、生成招投标日报/推送飞书

核心工作流程

1. 执行采集

cd "C:\Users\YC\.openclaw\workspace\skills\bidding-assistant"
python scripts/run_daily_collection.py --force

2. 生成PDF报告

# 生成月报
python scripts/generate_monthly_pdfs.py

# 生成日报
python scripts/pdf_generator.py --type daily --date 2026-05-04

3. 发送飞书通知

# 发送月报到飞书用户/群聊
python scripts/generate_and_send_monthly_report.py

数据库路径（重要！）

数据库：C:\Users\YC\.openclaw\workspace\skills\bidding-assistant\招投标数据\history.db

采集和PDF生成必须使用同一个数据库！

区域筛选规则

目标区域：盐南高新区、经开区
排除区域：亭湖、盐都、响水、滨海、建湖、大丰、射阳、阜宁、东台
筛选基于项目标题关键词匹配

支持的网站

盐城市政府采购网
开发区公共资源交易网
城南新区公共资源交易网
江苏世纪新城
盐城市大数据集团
盐城市东方集团
盐城市都市建设投资集团
经开城发集团
悦达集团
苏服采（API直连）

苏服采采集（API直连）

苏服采 https://js.fwgov.cn 通过直接调用后端API采集数据：

API: POST https://js.fwgov.cn:868/purchases/tenders/notice/page
地区代码: 盐南高新区=320992, 经开区=320991

飞书文件发送

使用IM API上传文件（不是Drive API）：

上传：POST /im/v1/files → 返回 file_key
发送：POST /im/v1/messages → content: {"file_key": xxx}

详细配置见 references/飞书企业自建应用配置指南.md

项目结构

bidding-assistant/
├── SKILL.md
├── assets/
│   └── image.png
├── references/
│   ├── 飞书企业自建应用配置指南.md
│   ├── 飞书推送配置说明.md
│   ├── PDF报告排版说明.md
│   ├── PDF报告生成与飞书推送使用指南.md
│   ├── PDF显示优化说明.md
│   ├── openclaw-browser-relay-guide.md
│   ├── sufu-workflow.md
│   └── website-config.md
└── scripts/
    ├── crawler.py                    # 采集器基类和所有网站采集器
    ├── sufu_crawler_final.py         # 苏服采采集器（API直连版）
    ├── pdf_generator.py              # PDF日报/月报生成器
    ├── feishu_enterprise_app.py       # 飞书开放平台API客户端
    ├── feishu_notifier.py            # 飞书Webhook通知器
    ├── generate_and_send_monthly_report.py  # 生成并发送月报
    ├── generate_monthly_pdfs.py      # 生成月报PDF
    ├── pdf_generator.py             # PDF生成器
    ├── run_daily_collection.py       # 每日采集任务入口
    └── scheduler.py                  # 定时任务调度器

依赖

requests>=2.28.0
beautifulsoup4>=4.11.0
lxml>=4.9.0
reportlab>=3.6.0
openpyxl>=3.1.0
apscheduler>=3.10.0

如使用浏览器自动化方式（苏服采旧版），还需：playwright>=1.35.0 和 playwright install chromium

安全使用建议

Before installing, confirm you trust the local Python scripts, install dependencies from trusted sources, use least-privilege Feishu credentials, verify report recipients, and decide whether you want the optional scheduler and local historical database enabled.

功能分析

Type: OpenClaw Skill Name: bidding-assistant Version: 1.0.0 The bidding-assistant skill bundle is a functional tool designed to automate the collection of bidding information from various Chinese government and corporate websites (e.g., czj.yancheng.gov.cn, js.fwgov.cn). The code utilizes standard libraries such as requests, BeautifulSoup, and Playwright for web scraping, and ReportLab for generating PDF reports. It integrates with the Feishu (Lark) API to send notifications and files, correctly retrieving credentials from the standard OpenClaw configuration path (~/.openclaw/openclaw.json). While the bundle contains hardcoded absolute file paths (e.g., C:\Users\YC\...) in SKILL.md and generate_monthly_pdfs.py, which suggests it was exported from a specific local environment without sanitization, the logic is entirely consistent with its stated purpose. No evidence of malicious intent, data exfiltration, or unauthorized execution was found.

能力评估

✓ Purpose & Capability

The artifacts consistently describe and implement collecting public bidding information, filtering by 盐南高新区/经开区, generating PDFs, and optionally pushing reports to Feishu.

ℹ Instruction Scope

The SKILL.md explicitly instructs running local Python scripts for collection, PDF generation, and Feishu sending; this is expected for the purpose but should be user-directed.

ℹ Install Mechanism

There is no install spec and no derived capability tags, while SKILL.md lists Python dependencies; users will need to manage dependency installation and provenance themselves.

ℹ Credentials

The skill performs network access to bidding sites and Feishu, writes a local SQLite database/PDFs/logs, and reads Feishu configuration when sending reports; these are purpose-aligned but sensitive enough to review.

ℹ Persistence & Privilege

It stores historical crawl data locally and includes an optional blocking scheduler for recurring daily tasks, but no artifact shows automatic startup or hidden persistence.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install bidding-assistant
安装完成后，直接呼叫该 Skill 的名称或使用 /bidding-assistant 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

第一次发布

元数据

Slug bidding-assistant

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题