webcatch

<div align="center">

# 🔍 Nexus-Insight

### 垂直领域商业舆情深度分析 Agent

**Vertical Domain Business Sentiment Analysis Agent**

[![Python](https://img.shields.io/badge/Python-3.10+-blue.svg)](https://www.python.org/)
[![Playwright](https://img.shields.io/badge/Playwright-Latest-green.svg)](https://playwright.dev/)
[![License](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)

</div>

---

> **声明 | Disclaimer**：本项目由温州大学生科院学生自主开发，旨在通过 AI 技术解决中小企业在非标品选品中的信息滞后问题。
> 
> This project is independently developed by a student from Wenzhou University, aiming to solve the information lag problem for SMEs in non-standard product selection through AI technology.

---

## 📖 项目简介 | Introduction

[中文](#中文) | [English](#english)

### English

Nexus-Insight is more than just a web crawler. It's a **Business Intelligence Agent** that integrates **Playwright** automation with **DeepSeek** analytical capabilities.

By simulating real user behavior, it deeply mines "unstructured reviews" from platforms like Xiaohongshu and Weibo, transforming them into quantifiable "business insights."

### 中文

Nexus-Insight 不仅仅是一个爬虫，它是一个集成 **Playwright** 自动化抓取与 **DeepSeek** 逻辑分析能力的商业情报 Agent。

它通过模拟真实用户行为，深入挖掘小红书、微博等平台的"非结构化评价"，并将其转化为可量化的"商业洞察"。

---

## ✨ 核心功能 | Core Features

| 功能 | Feature | 描述 |
|------|---------|------|
| 🕷️ **深度数据采集** | **Deep Data Collection** | 利用 Playwright 绕过动态加载限制，抓取实时舆情 |
| 🧠 **多维度情感分析** | **Multi-dimensional Sentiment Analysis** | 基于 LLM 对用户反馈进行分词处理与情感归类 |
| 📊 **自动化研报生成** | **Automated Report Generation** | 一键导出 Markdown 格式的深度研报 |
| 🕵️ **无痕巡航模式** | **Stealth Mode** | 采用隐私浏览器模式，规避算法推荐 |

---

## 🛠️ 技术栈 | Tech Stack

<div align="center">

| Category | Technology |
|----------|------------|
| 💻 Language | Python 3.10+ |
| 🔧 Framework | LangGraph, LangChain |
| 🌐 Browser | Playwright |
| 🤖 AI Model | DeepSeek-V3 / GPT-4o |

</div>

---

## 📦 快速开始 | Quick Start

### 1. 克隆仓库 | Clone Repository

```bash
git clone https://github.com/dsadsasdaddas/webcatch.git
cd web