Merge pull request #184 from jack6249/feature/gift-genius-agent

[毕业设计] GiftGeniusAgent - 智能送礼助手

Co-creation-projects/jack6249-GiftGeniusAgent/.env example

@@ -0,0 +1,8 @@
+#大模型参数
+LLM_MODEL_ID = "yourmodel"
+LLM_API_KEY = "yourkey"
+LLM_BASE_URL = "yourbaseurl"
+#Tavily参数
+TAVILY_API_KEY = "yourTavilyKey"
+#百度优选MCP参数
+BAIDU_MCP_TOKEN = "yourtoken"

Co-creation-projects/jack6249-GiftGeniusAgent/README.md

@@ -0,0 +1,143 @@
+# 🎁 GiftGenius: 智能送礼助手
+
+基于 HelloAgents 框架的多智能体协作系统，为你提供精准、走心的礼物推荐方案。
+
+## 📝 项目简介
+
+GiftGenius 是一个智能化的礼物推荐 Agent，旨在解决“送什么礼物”这个千古难题。它不仅仅是一个简单的关键词搜索工具，而是一个模拟人类决策过程的多智能体流水线 (Multi-Agent Pipeline)。
+
+通过 军师 (策略制定) -> 猎人 (全网搜索) -> 编辑 (数据清洗与文案创作) 的分工协作，它能根据用户的 MBTI、星座、预算等个性化画像，从全网检索最新的商品信息，并生成一份图文并茂、价格透明的送礼指南。
+
+- 解决什么问题？
+
+   解决送礼时的选择困难症，以及推荐商品过时、价格超预算、文案枯燥等问题。
+
+- 有什么特色功能？
+
+   支持 MBTI/星座心理分析、自动比价与平替查找、防幻觉数据提取。
+
+- 适用于什么场景？ 
+
+  节日送礼、生日惊喜、纪念日策划等需要个性化推荐的场景。
+
+## ✨ 核心功能
+
+- [x] 精准画像分析：基于 MBTI 人格、星座、年龄等维度，深度解析受礼者的潜在偏好，制定个性化搜索策略。
+
+- [x] 智能预算控制：支持自定义预算范围（如 "500-1000元"），并具备“价格守门员”机制，自动拦截超预算商品并触发降级搜索（找平替）。
+
+- [x]  实时联网搜索：利用 Tavily 搜索引擎获取 2025年最新 的商品信息、价格和图片，拒绝过时推荐。
+
+- [x] 可视化报告：最终生成包含商品图、价格参考、种草文案的 Markdown 表格，直观易读。
+
+## 🛠️ 技术栈
+
+- 框架: HelloAgents
+
+- 智能体范式: 使用HelloAgent框架的SimpleAgent
+
+- 工具与API:
+
+  Tavily Search API (用于联网检索)、百度优选MCP(用于联网检索)
+
+- 其他依赖: `mcp`, `nest_asyncio`, `python-dotenv`, `numpy`
+
+## 🚀 快速开始
+
+### 环境要求
+
+Python 3.10+
+
+Jupyter Notebook / Jupyter Lab
+
+### 安装依赖
+
+```bash
+pip install -r requirements.txt
+```
+
+### 配置API密钥
+
+复制配置文件模板：
+
+```bash
+# 创建.env文件
+cp .env.example .env
+# 编辑.env文件，填入你的API密钥
+```
+
+
+
+### 运行项目
+
+修改 user_profile.json 文件，填入你的送礼对象信息（如 MBTI、预算等）。
+
+启动 Jupyter Notebook：
+
+```bash
+jupyter notebook main.ipynb
+```
+项目默认使用的是百度优选MCP，可修改为Tavily Search API。
+```py
+# 搜索源配置
+# 可选值: "tavily" (通用/海外) 或 "baidu" (电商/国内)
+os.environ["SEARCH_PROVIDER"] = "baidu" 
+```
+
+
+点击 "Run All" 运行所有单元格，最终结果将生成在 outputs/gift_plan_output.md 中。
+
+## 📖 使用示例
+
+输入配置 (user_profile.json):
+
+```json
+{
+    "性别": "男",
+    "年龄": "24岁",
+    "MBTI": "ISTJ",
+    "星座": "白羊座",
+    "预算": "200-500",
+    "节日": "生日",
+    "自定义": "喜欢数码"
+}
+```
+
+运行结果 (final_gift_plan.md):
+
+![example](https://github.com/datawhalechina/hello-agents/blob/main/Co-creation-projects/jack6249-GiftGeniusAgent/example.png)
+
+## 🎯 项目亮点
+
+- 双流架构 (Dual-Stream)：将“硬数据搜索”（找价格）和“软文案生成”（找卖点）拆分为两条并行流水线，大幅减少了上下文干扰，提升了文案质量。
+
+- 代码级防幻觉 (Code-based Guardrails)：不依赖 LLM 直接生成 JSON，而是通过 Python 正则表达式从搜索结果中暴力提取价格和图片，从根源上杜绝了“编造价格”的幻觉。
+
+- 动态策略修正 (Feedback Loop)：实现了“价格守门员”机制。如果搜到的商品均价超预算，会重新触发“军师”制定“平替”策略，直到找到合适商品为止。
+
+- 支持多数据源：集成了百度优选MCP 和 Tavily Search API 两种搜索源
+
+## 🔮 未来计划
+
+- [ ] 前端交互：新增前端页面，提供更好的用户交互体验
+
+- [ ] 数据源深度集成：完全接入百度优选MCP 的比价与历史价格接口，获取更精准的实时价格和库存信息，实现“全网比价”功能。
+
+- [ ] 丰富选项：增加更多的个人喜好选项，如喜欢的商品类型、品牌等
+
+
+🤝 贡献指南
+
+欢迎提出 Issue 和 Pull Request！如果你有更好的 Prompt 优化技巧或新的 Agent 模式想法，请随时分享。
+
+📄 许可证
+
+MIT License
+
+👤 作者
+
+GitHub: [@jack6249](https://github.com/jack6249)
+
+🙏 致谢
+
+感谢 Datawhale 社区 和 Hello-Agents 项目提供的优秀框架与教程支持！

Co-creation-projects/jack6249-GiftGeniusAgent/data/test_cases.json

@@ -0,0 +1,9 @@
+{
+    "性别": "男",
+    "年龄": "24岁",
+    "MBTI": "ISTJ",
+    "星座": "白羊座",
+    "预算": "200-500",
+    "节日": "生日",
+    "自定义": "喜欢数码"
+}

Co-creation-projects/jack6249-GiftGeniusAgent/main.ipynb

@@ -0,0 +1,725 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "38b007c8",
+   "metadata": {},
+   "source": [
+    "\n",
+    "## GiftGeniusAgent——你的送礼智能Agent\n",
+    "\n",
+    "### 项目简介\n",
+    "本项目演示一个基于HelloAgents框架的智能送礼Agent\n",
+    "\n",
+    "### 作者信息\n",
+    "- 姓名：张善祺\n",
+    "- GitHub:@jack6249\n",
+    "- 日期：2025-11-21\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "62a3b6a3",
+   "metadata": {},
+   "source": [
+    "### 第1部分：环境配置"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "dd9b8a20",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#导入库和参数配置\n",
+    "from hello_agents import SimpleAgent, HelloAgentsLLM, ReflectionAgent, ToolRegistry\n",
+    "from hello_agents.tools import Tool, ToolParameter\n",
+    "from typing import Dict, Any, List\n",
+    "from tavily import TavilyClient\n",
+    "import os\n",
+    "import json\n",
+    "import re\n",
+    "import numpy as np \n",
+    "from dotenv import load_dotenv\n",
+    "import asyncio\n",
+    "import nest_asyncio\n",
+    "from mcp.client.sse import sse_client\n",
+    "from mcp.client.session import ClientSession\n",
+    "\n",
+    "load_dotenv()\n",
+    "\n",
+    "#LLM参数\n",
+    "LLM_MODEL_ID = os.getenv(\"LLM_MODEL_ID\")\n",
+    "LLM_API_KEY = os.getenv(\"LLM_API_KEY\")\n",
+    "LLM_BASE_URL = os.getenv(\"LLM_BASE_URL\")\n",
+    "#Tavily参数\n",
+    "TAVILY_API_KEY = os.getenv(\"TAVILY_API_KEY\",\"\")\n",
+    "#百度MCP参数\n",
+    "BAIDU_TOKEN = os.getenv(\"BAIDU_MCP_TOKEN\",\"\")\n",
+    "#输入json路径配置\n",
+    "INPUT_FILENAME = \"data/test_cases.json\"\n",
+    "\n",
+    "# 搜索源配置\n",
+    "# 可选值: \"tavily\" (通用/海外) 或 \"baidu\" (电商/国内)\n",
+    "os.environ[\"SEARCH_PROVIDER\"] = \"baidu\" \n",
+    "\n",
+    "print(\"✅ 环境配置完成\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "9404c299",
+   "metadata": {},
+   "source": [
+    "### 第2部分：定义工具"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "37b994bc",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# [Cell 2 终极版] 定义统一搜索工具 (兼容 Tavily 和 Baidu)\n",
+    "# 允许 Jupyter 运行异步\n",
+    "nest_asyncio.apply()\n",
+    "\n",
+    "class BatchSearchTool(Tool):\n",
+    "    def __init__(self):\n",
+    "        super().__init__(\n",
+    "            name=\"batch_search\",\n",
+    "            description=\"统一搜索工具，支持 Tavily 和 Baidu 切换。\"\n",
+    "        )\n",
+    "        self.provider = os.environ.get(\"SEARCH_PROVIDER\", \"tavily\").lower()\n",
+    "\n",
+    "    def run(self, parameters: Any) -> str:\n",
+    "        return \"请使用 Python 代码直接调用 search_raw 方法获取数据。\"\n",
+    "\n",
+    "    def search_raw(self, query: str) -> List[Dict]:\n",
+    "        if self.provider == \"baidu\":\n",
+    "            return self._search_baidu(query)\n",
+    "        else:\n",
+    "            return self._search_tavily(query)\n",
+    "\n",
+    "    # --- 引擎 A: Tavily ---\n",
+    "    def _search_tavily(self, query: str) -> List[Dict]:\n",
+    "        api_key = os.environ.get(\"TAVILY_API_KEY\")\n",
+    "        if not api_key: return []\n",
+    "        print(f\"    🚀 [Tavily] 正在搜索: {query} ...\")\n",
+    "        try:\n",
+    "            tavily = TavilyClient(api_key=api_key)\n",
+    "            response = tavily.search(query, max_results=5, include_images=True)\n",
+    "            results = []\n",
+    "            if 'results' in response:\n",
+    "                for r in response['results']:\n",
+    "                    results.append({\n",
+    "                        \"title\": r['title'], \"url\": r['url'], \"content\": r['content'], \n",
+    "                        \"type\": \"text\", \"img\": \"\" # Tavily 文本通常不带图\n",
+    "                    })\n",
+    "            if 'images' in response and response['images']:\n",
+    "                results.append({\"images\": response['images'][:3], \"type\": \"image\"})\n",
+    "            return results\n",
+    "        except Exception as e:\n",
+    "            print(f\"      ⚠️ Tavily 异常: {e}\")\n",
+    "            return []\n",
+    "\n",
+    "    # --- 引擎 B: Baidu MCP ---\n",
+    "    def _search_baidu(self, query: str) -> List[Dict]:\n",
+    "        token = os.environ.get(\"BAIDU_MCP_TOKEN\")\n",
+    "        if not token: return []\n",
+    "        print(f\"    🐼 [百度优选] 正在搜索: {query} ...\")\n",
+    "        try:\n",
+    "            raw_json_str = asyncio.run(self._async_baidu_call(query, token))\n",
+    "            print(f\"      🔍 原始 JSON 响应: {raw_json_str}\")\n",
+    "            return self._parse_baidu_response(raw_json_str)\n",
+    "        except Exception as e:\n",
+    "            print(f\"      ⚠️ 百度 MCP 异常: {e}\")\n",
+    "            return []\n",
+    "\n",
+    "    async def _async_baidu_call(self, query: str, token: str) -> str:\n",
+    "        sse_url = f\"https://mcp-youxuan.baidu.com/mcp/sse?key={token}\"\n",
+    "        async with sse_client(sse_url) as (read, write):\n",
+    "            async with ClientSession(read, write) as session:\n",
+    "                await session.initialize()\n",
+    "                result = await session.call_tool(\"goods_search\", arguments={\"query\": query})\n",
+    "                return result.content[0].text if result.content else \"\"\n",
+    "\n",
+    "    def _parse_baidu_response(self, json_str: str) -> List[Dict]:\n",
+    "        results = []; images = []\n",
+    "        try:\n",
+    "            data = json.loads(json_str)\n",
+    "            items = data if isinstance(data, list) else []\n",
+    "            \n",
+    "            for item in items[:5]:\n",
+    "                title = item.get(\"goodsName\") or item.get(\"title\") or \"未知商品\"\n",
+    "                price = item.get(\"price\") or item.get(\"minPrice\") or \"\"\n",
+    "                shop = item.get(\"shopName\") or item.get(\"mall\") or \"\"\n",
+    "                url = item.get(\"detailUrl\") or item.get(\"url\") or item.get(\"ori_url\") or \"#\"\n",
+    "                img = item.get(\"imgUrl\") or item.get(\"picUrl\") or item.get(\"img\")\n",
+    "                \n",
+    "                content = f\"价格: {price}元。店铺: {shop}。商品详情: {title}\"\n",
+    "                \n",
+    "                # 📝【修复点】直接在 text 类型结果里绑定 img\n",
+    "                results.append({\n",
+    "                    \"title\": title, \"url\": url, \"content\": content, \n",
+    "                    \"type\": \"text\", \"img\": img \n",
+    "                })\n",
+    "                if img: images.append(img)\n",
+    "            \n",
+    "            if images: results.append({\"images\": images[:3], \"type\": \"image\"})\n",
+    "                \n",
+    "        except json.JSONDecodeError:\n",
+    "            print(\"      ⚠️ 百度返回非 JSON 数据\")\n",
+    "        return results\n",
+    "\n",
+    "    def get_parameters(self):\n",
+    "        return [ToolParameter(name=\"query\", type=\"string\", description=\"关键词\")]\n",
+    "\n",
+    "tool_registry = ToolRegistry()\n",
+    "tool_registry.register_tool(BatchSearchTool())\n",
+    "\n",
+    "print(\"✅ 统一搜索工具已加载！\")\n",
+    "print(f\"当前模式: {'百度优选 (电商)' if os.environ.get('SEARCH_PROVIDER') == 'baidu' else 'Tavily (通用)'}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e8ffecb0",
+   "metadata": {},
+   "source": [
+    "### 第3部分：创建智能体\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "6c5a2e82",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# 初始化大模型\n",
+    "llm = HelloAgentsLLM()\n",
+    "\n",
+    "# --- 1. 军师 (Profiler) - 已升级支持多维度画像 ---\n",
+    "PROFILER_PROMPT = \"\"\"\n",
+    "你是一个精通 MBTI 人格分析与消费市场趋势的 \"送礼军师\"。\n",
+    "你的任务是根据用户提供的多维度画像，制定 3 个**极度精准**的搜索关键词。\n",
+    "\n",
+    "【⚠️ 时效性死命令 (CRITICAL)】\n",
+    "当前时间视作 **2025年11月**。\n",
+    "1. **严禁过时**：绝对不要推荐 2024 年或更早的旧款（除非是经典恒久款如黑胶唱片）。\n",
+    "2. **价格尺度**：给出的价格单位是人民币元。请严格遵照范围进行联想，禁止超出预算范围。\n",
+    "\n",
+    "为了确保推荐质量，请参考以下的【优秀思考范例】：\n",
+    "\n",
+    "### 范例 1\n",
+    "**用户画像**: \n",
+    "- 女, 26岁, ISFP (探险家), 金牛座\n",
+    "- 预算: 500-1000元\n",
+    "- 场景: 情人节\n",
+    "- 自定义: 喜欢有质感的生活小物\n",
+    "**军师分析**: \n",
+    "ISFP 重视审美和感官体验，金牛座喜欢实实在在的质感。情人节需要浪漫。\n",
+    "**生成策略**:\n",
+    "1. 观夏 (To Summer) 昆仑煮雪 晶石香薰 (符合质感与审美)\n",
+    "2. 野兽派 2025 情人节限定 睡衣礼盒 (金牛座喜欢的舒适)\n",
+    "3. 富士 Instax mini Evo 拍立得 (记录生活瞬间)\n",
+    "\n",
+    "### 范例 2\n",
+    "**用户画像**: \n",
+    "- 男, 30岁, INTJ (建筑师), 处女座\n",
+    "- 预算: 1000元以上\n",
+    "- 场景: 生日\n",
+    "- 自定义: 程序员，喜欢整洁\n",
+    "**军师分析**: \n",
+    "INTJ 追求极致的逻辑和效率，处女座有洁癖，喜欢桌面整洁。\n",
+    "**生成策略**:\n",
+    "1. Keychron Q1 Pro 机械键盘 铝坨坨 (符合极客对工具的追求)\n",
+    "2. 明基 (BenQ) ScreenBar Halo 屏幕挂灯 (极致护眼与桌面美学)\n",
+    "3. 赫曼米勒 (Herman Miller) 显示器支架 (人体工学)\n",
+    "\n",
+    "### 范例 3\n",
+    "**用户画像**: \n",
+    "- 女, 20岁, ENFP (竞选者), 狮子座\n",
+    "- 预算: 300元以内\n",
+    "- 场景: 圣诞节\n",
+    "- 自定义: 喜欢二次元，痛包\n",
+    "**军师分析**: \n",
+    "ENFP 热情奔放，狮子座喜欢张扬、闪亮的东西。预算有限但要素多。\n",
+    "**生成策略**:\n",
+    "1. 泡泡玛特 圣诞系列 盲盒整端 (符合节日气氛和二次元)\n",
+    "2. WEGO 痛包 镭射款 (符合自定义需求，狮子座喜欢的亮眼)\n",
+    "3.  Chiikawa 吉伊卡哇 圣诞公仔 (当下顶流二次元IP)\n",
+    "\n",
+    "---\n",
+    "\n",
+    "**现在的任务**：\n",
+    "请根据以下【当前用户画像】进行分析，模仿上述范例的深度，制定搜索策略。\n",
+    "\n",
+    "【当前用户画像】\n",
+    "{user_profile_text}\n",
+    "\n",
+    "【关键词生成要求】\n",
+    "1. **必须具体**：格式为 `[品牌] + [产品名/系列] + [限定/属性]`。\n",
+    "2. **拒绝大词**：严禁搜索 \"礼物\"、\"口红\"、\"玩具\" 这种泛词。\n",
+    "3. **必须包含品牌**：根据预算推断合适的品牌（如：预算低选名创优品/泡泡玛特，预算高选Dior/索尼）。\n",
+    "\n",
+    "【输出格式】\n",
+    "只输出 3 行搜索关键词，每行一个。不要输出分析过程，不要序号。\n",
+    "\"\"\"\n",
+    "\n",
+    "profiler_agent = SimpleAgent(\n",
+    "    llm=llm,\n",
+    "    name=\"Agent_Profiler\",\n",
+    "    system_prompt=PROFILER_PROMPT\n",
+    ")\n",
+    "\n",
+    "# ==============================================================================\n",
+    "# 2. 种草达人 (Pitcher) - 文案创作 (加入风格指导)\n",
+    "# ==============================================================================\n",
+    "PITCHER_PROMPT = \"\"\"\n",
+    "你是一个 **金牌种草文案**。\n",
+    "用户会给你一个 **【商品名称】**。\n",
+    "\n",
+    "### 🎯 关键要点 (Few Points)\n",
+    "1.  **痛点直击**: 一句话说清楚为什么买它（限定？显白？绝美？）。\n",
+    "2.  **情绪价值**: 使用 \"绝绝子\", \"氛围感\", \"心动\" 等高频热词。\n",
+    "3.  **字数限制**: 严格控制在 **40字以内**，短小精悍。\n",
+    "4.  **Emoji**: 必须包含 1-2 个 emoji。\n",
+    "\n",
+    "### 🌟 创作范例 (Few-Shot)\n",
+    "**输入**: Dior 999 烈艳蓝金\n",
+    "**输出**: 💄本宫不死终是妃！Dior 999 传奇正红，显白更有气场，送女友绝对没错！\n",
+    "\n",
+    "**输入**: 泡泡玛特 Labubu 坐坐派对\n",
+    "**输出**: ✨太可爱了吧！Labubu 坐坐派对系列，每一个都丑萌到心巴上，摆在桌上超治愈~\n",
+    "\n",
+    "**输入**: 罗技 MX Master 3S\n",
+    "**输出**: 🖱️打工人本命！罗技 Master 3S 静音又顺滑，人体工学设计，手腕再也不累了。\n",
+    "\n",
+    "---\n",
+    "\n",
+    "**当前任务**：\n",
+    "请为【{input}】写一句朋友圈风格种草语。\n",
+    "\"\"\"\n",
+    "pitcher_agent = SimpleAgent(\n",
+    "    llm=llm,\n",
+    "    name=\"Agent_Pitcher\",\n",
+    "    system_prompt=PITCHER_PROMPT\n",
+    ")\n",
+    "\n",
+    "print(\"✅ 智能体初始化完成！\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "25907c14",
+   "metadata": {},
+   "source": [
+    "### 第4部分:读取数据"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "58bd39dd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def load_user_profile(filename):\n",
+    "    # 1. 检查文件是否存在\n",
+    "    if not os.path.exists(filename):\n",
+    "        print(f\"⚠️ 未找到配置文件: {filename}\")\n",
+    "        # 如果没有文件，将默认数据写入文件\n",
+    "        default_data = {\n",
+    "            \"性别\": \"女\",\n",
+    "            \"年龄\": \"24岁\",\n",
+    "            \"MBTI\": \"ENFP\",\n",
+    "            \"星座\": \"天秤座\",\n",
+    "            \"预算\": \"500元以内\",\n",
+    "            \"节日\": \"恋爱一周年纪念日\",\n",
+    "            \"自定义\": \"喜欢二次元，平时喜欢喝咖啡，不要送太实用的家电\"\n",
+    "        }\n",
+    "        with open(filename, \"w\", encoding=\"utf-8\") as f:\n",
+    "            json.dump(default_data, f, ensure_ascii=False, indent=4)\n",
+    "        print(f\"✅ 已自动生成默认配置文件，请修改 {filename} 后再次运行。\")\n",
+    "        return default_data\n",
+    "\n",
+    "    # 2. 读取文件内容\n",
+    "    try:\n",
+    "        with open(filename, \"r\", encoding=\"utf-8\") as f:\n",
+    "            data = json.load(f)\n",
+    "        print(f\"✅ 成功加载用户画像: {filename}\")\n",
+    "        print(f\"📋 内容预览: {json.dumps(data, ensure_ascii=False)}\")\n",
+    "        return data\n",
+    "    except Exception as e:\n",
+    "        print(f\"❌ 读取 JSON 失败: {e}\")\n",
+    "        return {}\n",
+    "\n",
+    "# 加载数据\n",
+    "user_input_data = load_user_profile(INPUT_FILENAME)\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "0fdcab5a",
+   "metadata": {},
+   "source": [
+    "### 第5部分：生成礼物计划"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "cdaef6b1",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def parse_budget_range(budget_str):\n",
+    "    \"\"\"解析用户预算字符串，返回 (min, max)\"\"\"\n",
+    "    nums = [float(x) for x in re.findall(r'\\d+', str(budget_str).replace(',', ''))]\n",
+    "    if not nums: return 0, 999999 \n",
+    "    if \"以内\" in budget_str or \"以下\" in budget_str: return 0, nums[0]\n",
+    "    if \"以上\" in budget_str: return nums[0], 999999\n",
+    "    if len(nums) >= 2: return min(nums), max(nums)\n",
+    "    return 0, nums[0]\n",
+    "\n",
+    "def extract_all_prices(raw_results):\n",
+    "    \"\"\"从搜索结果列表中提取所有有效的价格\"\"\"\n",
+    "    prices = []\n",
+    "    for res in raw_results:\n",
+    "        # 只处理文本类型的结果\n",
+    "        if res.get('type') == 'text':\n",
+    "            text = res.get('title', '') + \" \" + res.get('content', '')\n",
+    "            # 匹配 ¥, $, 元 等格式\n",
+    "            matches = re.findall(r'(?:¥|￥|\\$|HK\\$|NT\\$)\\s*(\\d+(?:,\\d{3})*(?:\\.\\d+)?)', text)\n",
+    "            for m in matches:\n",
+    "                val = float(m.replace(',', ''))\n",
+    "                # 过滤掉像年份(2025)或过小/过大的异常值\n",
+    "                if 10 < val < 100000 and val not in [2024, 2025, 2026]:\n",
+    "                    prices.append(val)\n",
+    "            # 备用正则：匹配 \"xxx元\"\n",
+    "            matches_yuan = re.findall(r'(\\d+(?:,\\d{3})*(?:\\.\\d+)?)\\s*元', text)\n",
+    "            for m in matches_yuan:\n",
+    "                val = float(m.replace(',', ''))\n",
+    "                if 10 < val < 100000 and val not in [2024, 2025, 2026]:\n",
+    "                    prices.append(val)\n",
+    "    return prices\n",
+    "\n",
+    "\n",
+    "def find_best_product(hunter, profiler_agent, keyword, budget_min, budget_max):\n",
+    "    limit_upper = budget_max * 1.2\n",
+    "    limit_lower = budget_min * 0.8\n",
+    "    \n",
+    "    all_candidates = []\n",
+    "    current_kw = keyword\n",
+    "    \n",
+    "    # --- Round 1: 首次搜索 ---\n",
+    "    print(f\"       🕵️ 第1次搜索: {current_kw} 价格\")\n",
+    "    results_1 = hunter.search_raw(f\"{current_kw} 价格\")\n",
+    "    \n",
+    "    fallback_img = \"\"\n",
+    "    for r in results_1:\n",
+    "        if r.get('images'): \n",
+    "            fallback_img = r['images'][0]\n",
+    "            break\n",
+    "\n",
+    "    has_valid_info = False\n",
+    "    for res in results_1:\n",
+    "        if res.get('type') == 'text':\n",
+    "            has_valid_info = True\n",
+    "            p_vals = extract_all_prices([res])\n",
+    "            if p_vals:\n",
+    "                res['price_val'] = p_vals[0]\n",
+    "                # 📝【核心修复点1】记录当前结果所属的关键词\n",
+    "                res['source_kw'] = current_kw \n",
+    "                all_candidates.append(res)\n",
+    "                \n",
+    "                if limit_lower <= p_vals[0] <= limit_upper:\n",
+    "                    if not res.get('img') and fallback_img: res['img'] = fallback_img\n",
+    "                    return res, f\"约 {p_vals[0]}元\", current_kw\n",
+    "\n",
+    "    # --- 机制 3: 无数据防御 ---\n",
+    "    if not has_valid_info:\n",
+    "        print(f\"       ⚠️ [机制3触发] 首次搜索无有效信息。\")\n",
+    "        correction_prompt = f\"原策略 '{current_kw}' 搜索结果为空，请推荐一个同品类但更热门的具体商品型号。只输出关键词。\"\n",
+    "        new_kw = profiler_agent.run(correction_prompt).strip()\n",
+    "        print(f\"       🔄 军师换词: {new_kw}\")\n",
+    "        current_kw = new_kw\n",
+    "        \n",
+    "        results = hunter.search_raw(f\"{current_kw} 价格\")\n",
+    "        \n",
+    "        # 更新 fallback_img\n",
+    "        fallback_img = \"\" \n",
+    "        for r in results:\n",
+    "            if r.get('images'): \n",
+    "                fallback_img = r['images'][0]\n",
+    "                break\n",
+    "                \n",
+    "        for res in results:\n",
+    "            if res.get('type') == 'text':\n",
+    "                p_vals = extract_all_prices([res])\n",
+    "                if p_vals:\n",
+    "                    res['price_val'] = p_vals[0]\n",
+    "                    # 📝【核心修复点1】记录关键词\n",
+    "                    res['source_kw'] = current_kw\n",
+    "                    all_candidates.append(res)\n",
+    "\n",
+    "    # --- 机制 1 & 2: 价格修正 ---\n",
+    "    avg_price = np.mean([c['price_val'] for c in all_candidates]) if all_candidates else 0\n",
+    "    \n",
+    "    if avg_price > 0:\n",
+    "        correction_prompt = \"\"\n",
+    "        if avg_price > limit_upper:\n",
+    "            print(f\"       💸 [机制1触发] 均价 {int(avg_price)} > 上限 {int(limit_upper)}，找平替...\")\n",
+    "            correction_prompt = f\"原策略 '{current_kw}' 均价约 {int(avg_price)}元，超预算 ({budget_max}元)。请推荐一个同品类更便宜的具体型号（平替）。只输出关键词。\"\n",
+    "        elif avg_price < limit_lower:\n",
+    "            print(f\"       📉 [机制2触发] 均价 {int(avg_price)} < 下限 {int(limit_lower)}，找升级款...\")\n",
+    "            correction_prompt = f\"原策略 '{current_kw}' 均价约 {int(avg_price)}元，低于预算下限 ({budget_min}元)。请推荐一个同品类更高端的型号。只输出关键词。\"\n",
+    "            \n",
+    "        if correction_prompt:\n",
+    "            new_kw = profiler_agent.run(correction_prompt).strip()\n",
+    "            print(f\"       🔄 军师修正: {new_kw}\")\n",
+    "            current_kw = new_kw\n",
+    "            \n",
+    "            results_2 = hunter.search_raw(f\"{new_kw} 价格\")\n",
+    "            \n",
+    "            # 更新 fallback_img\n",
+    "            fallback_img = \"\" \n",
+    "            for r in results_2:\n",
+    "                if r.get('images'): \n",
+    "                    fallback_img = r['images'][0]\n",
+    "                    break\n",
+    "            \n",
+    "            for res in results_2:\n",
+    "                if res.get('type') == 'text':\n",
+    "                    p_vals = extract_all_prices([res])\n",
+    "                    if p_vals:\n",
+    "                        res['price_val'] = p_vals[0]\n",
+    "                        # 📝【核心修复点1】记录关键词\n",
+    "                        res['source_kw'] = current_kw\n",
+    "                        all_candidates.append(res) \n",
+    "                        \n",
+    "                        if limit_lower <= p_vals[0] <= limit_upper:\n",
+    "                            if not res.get('img') and fallback_img: res['img'] = fallback_img\n",
+    "                            tag = \"(平替)\" if avg_price > limit_upper else \"(升级)\"\n",
+    "                            return res, f\"约 {p_vals[0]}元 {tag}\", current_kw\n",
+    "\n",
+    "    # --- 机制 4: 兜底防御 ---\n",
+    "    print(\"       ⚠️ [机制4触发] 启用强制兜底模式...\")\n",
+    "    best_fallback = None\n",
+    "    status_msg = \"暂无报价\"\n",
+    "    \n",
+    "    if all_candidates:\n",
+    "        # 选离预算最近的\n",
+    "        target = (budget_min + budget_max) / 2\n",
+    "        best_fallback = sorted(all_candidates, key=lambda x: abs(x['price_val'] - target))[0]\n",
+    "        p = best_fallback['price_val']\n",
+    "        \n",
+    "        if p > limit_upper: status_msg = f\"约 {p}元 (⚠️超预算)\"\n",
+    "        elif p < limit_lower: status_msg = f\"约 {p}元 (📉低于预算)\"\n",
+    "        else: status_msg = f\"约 {p}元\"\n",
+    "        \n",
+    "    elif results_1:\n",
+    "        # 实在没数据，硬取第一条\n",
+    "        for res in results_1:\n",
+    "            if res.get('type') == 'text': \n",
+    "                best_fallback = res\n",
+    "                # 兜底时如果也没价格，就用原始关键词\n",
+    "                best_fallback['source_kw'] = keyword \n",
+    "                break\n",
+    "    \n",
+    "    if best_fallback:\n",
+    "        if not best_fallback.get('img') and fallback_img:\n",
+    "            best_fallback['img'] = fallback_img\n",
+    "        \n",
+    "        # 📝【核心修复点2】返回结果里记录的那个 source_kw，而不是当前的 current_kw\n",
+    "        final_name_to_use = best_fallback.get('source_kw', current_kw)\n",
+    "        \n",
+    "        return best_fallback, status_msg, final_name_to_use\n",
+    "        \n",
+    "    return None, \"搜索失败\", keyword"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "4ec82cc7",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "if not user_input_data:\n",
+    "    print(\"❌ 未加载用户数据\")\n",
+    "else:\n",
+    "    # 0. 解析预算\n",
+    "    b_min, b_max = parse_budget_range(user_input_data.get('预算', ''))\n",
+    "    print(f\"\\n💰 预算范围: {b_min} - {b_max}元\")\n",
+    "\n",
+    "    # 1. 军师制定策略\n",
+    "    profile_text = \"\\n\".join([f\"- {k}: {v if v else '未知/不限'}\" for k, v in user_input_data.items()])\n",
+    "    print(f\"\\n🚀 任务启动...\\n{'-'*40}\")\n",
+    "    print(\"\\n🧠 [1/3] 军师正在制定初步策略...\")\n",
+    "    search_strategy = profiler_agent.run(f\"请根据以下用户画像制定搜索策略：\\n\\n{profile_text}\")\n",
+    "    print(f\"📝 策略: \\n{search_strategy}\")\n",
+    "\n",
+    "    # 2. 准备循环\n",
+    "    keywords = [k.strip() for k in search_strategy.replace(\"，\", \",\").replace(\"\\n\", \",\").split(',') if k.strip()]\n",
+    "    final_items = []\n",
+    "    hunter = BatchSearchTool()\n",
+    "\n",
+    "    print(f\"\\n🔄 进入处理流程 (共 {len(keywords)} 个商品)...\")\n",
+    "\n",
+    "    for index, kw in enumerate(keywords):\n",
+    "        print(f\"\\n    👉 [商品 {index+1}/{len(keywords)}] 正在处理: {kw}\")\n",
+    "        \n",
+    "        # 调用智能搜索函数 (传入 min 和 max)\n",
+    "        valid_result, price_status, final_kw = find_best_product(hunter, profiler_agent, kw, b_min, b_max)\n",
+    "        \n",
+    "        if not valid_result:\n",
+    "            print(\"       ❌ 彻底无数据，跳过。\")\n",
+    "            continue\n",
+    "            \n",
+    "        # === 生成文案 ===\n",
+    "        product_name = valid_result.get('title', final_kw)\n",
+    "        \n",
+    "        print(f\"       ✍️ 正在撰写文案: {product_name[:30]}...\")\n",
+    "        pitch_prompt = f\"\"\"\n",
+    "        商品：{product_name}\n",
+    "        价格：{price_status}\n",
+    "        卖点片段：{valid_result.get('content', '')[:200]}...\n",
+    "        \n",
+    "        请写一句30字以内的种草文案。\n",
+    "        \"\"\"\n",
+    "        pitch = pitcher_agent.run(pitch_prompt)\n",
+    "        \n",
+    "        final_items.append({\n",
+    "            \"name\": final_kw, \n",
+    "            \"title_full\": product_name,\n",
+    "            \"price\": price_status,\n",
+    "            \"desc\": pitch.replace(\"\\n\", \" \").strip(),\n",
+    "            \"img\": valid_result.get('img', ''),\n",
+    "            \"link\": valid_result.get('url', '')\n",
+    "        })\n",
+    "        print(f\"       ✅ 已收录 (状态: {price_status})\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "0197ef43",
+   "metadata": {},
+   "source": [
+    "\n",
+    "### 第6部分：输出礼物计划"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8426fb13",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# --- 4. 渲染与保存 ---\n",
+    "print(f\"\\n💾 正在生成最终报告...\")\n",
+    "\n",
+    "if not final_items:\n",
+    "    final_md = \"很抱歉，网络搜索似乎出现了问题，未能获取到任何商品信息。\"\n",
+    "else:\n",
+    "    table_header = \"| 🎁 礼物名称 | 💰 价格 | ✨ 种草理由 | 🖼️ 图片/链接 |\\n| :--- | :--- | :--- | :--- |\\n\"\n",
+    "    table_rows = []\n",
+    "    \n",
+    "    for item in final_items:\n",
+    "        # 1. 清洗文本字段 (防止 | 破坏表格)\n",
+    "        name = item.get('name', '未知').replace(\"|\", \"/\")\n",
+    "        price = item.get('price', '暂无').replace(\"|\", \"/\")\n",
+    "        desc = item.get('desc', '').replace(\"|\", \"/\")\n",
+    "        \n",
+    "        # 2. 🚨【核心修复】清洗链接中的竖线\n",
+    "        # 百度/京东链接常包含 '|'，必须替换为 '%7C'，否则 Markdown 表格会炸\n",
+    "        raw_link = item.get('link', '#')\n",
+    "        safe_link = raw_link.replace(\"|\", \"%7C\")\n",
+    "        \n",
+    "        raw_img = item.get('img', '')\n",
+    "        safe_img = raw_img.replace(\"|\", \"%7C\")\n",
+    "        \n",
+    "        # 3. 构建媒体列\n",
+    "        if safe_img and safe_img.startswith(\"http\"):\n",
+    "            # 图片链接套购买链接\n",
+    "            media = f\"[![图]({safe_img})]({safe_link})\"\n",
+    "        else:\n",
+    "            media = f\"[点击购买]({safe_link})\"\n",
+    "        \n",
+    "        # 4. 组装行 (注意名字上的链接也要用 safe_link)\n",
+    "        # 使用 strip() 去除可能的首尾空格\n",
+    "        row = f\"| [{name}]({safe_link}) | {price} | {desc} | {media} |\"\n",
+    "        table_rows.append(row)\n",
+    "        \n",
+    "    final_md = table_header + \"\\n\".join(table_rows)\n",
+    "filename = \"outputs/gift_plan_output.md\"\n",
+    "# 确保输出目录存在\n",
+    "os.makedirs(os.path.dirname(filename), exist_ok=True)\n",
+    "\n",
+    "with open(filename, \"w\", encoding=\"utf-8\") as f:\n",
+    "    f.write(final_md)\n",
+    "print(f\"🎉 任务完成！文件已保存: {os.path.abspath(filename)}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d8b884b5",
+   "metadata": {},
+   "source": [
+    "### 第7部分：总结与展望"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "08b0b8e8",
+   "metadata": {},
+   "source": [
+    "#### 实现的功能\n",
+    "- 基于用户输入的个人信息，生成符合预算的礼物建议\n",
+    "- 支持用户自定义预算范围、节日、个人喜好等\n",
+    "- 支持百度MCP和Tavily API双数据源，利用搜索引擎获取最新的商品信息和价格\n",
+    "- 提供可视化的建议结果展示\n",
+    "#### 遇到的挑战与解决方案\n",
+    "- 大模型的“幻觉”问题（JSON格式错误/编造数据）\n",
+    "  - 解决方案：放弃让 LLM 直接生成最终数据。改为使用 Python 正则表达式 从搜索结果中暴力提取硬数据（价格、图片），仅让 LLM 负责生成文案。代码逻辑负责准确性，模型负责创造性。\n",
+    "- 上下文过长导致提取失败\n",
+    "  - 解决方案：结合实际业务场景，分析各个阶段对上下文的要求，在搜索阶段限制返回长度。同时通过拆分 “硬数据流”（找参数）和 “软数据流”（找卖点），大幅降低单次上下文长度，提升响应速度。\n",
+    "- 大模型推荐的礼品价格超出预算\n",
+    "  - 解决方案：引入检核机制。如果搜到的商品均价超预算，系统会自动呼叫“军师”重新制定“平替”策略，直到找到合适商品为止。\n",
+    "- Agent传入的参数格式问题\n",
+    "  - 解决方案：在工具层兼容 Agent 传入的各种参数格式（JSON/字符串、逗号/换行符分隔），确保搜索指令不丢失。\n",
+    "#### 未来改进方向\n",
+    "- 前端交互：开发前端页面，替代目前的Notebook交互，提供更好的用户交互体验\n",
+    "- 数据源深度集成：完全接入百度优选MCP 的比价与历史价格接口，获取更精准的实时价格和库存信息，实现“全网比价”\n",
+    "- 丰富选项：增加更多的个人喜好选项，如喜欢的商品类型、品牌等\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "ai_3.10",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.19"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

Co-creation-projects/jack6249-GiftGeniusAgent/outputs/gift_plan_output.md

@@ -0,0 +1,5 @@
+| 🎁 礼物名称 | 💰 价格 | ✨ 种草理由 | 🖼️ 图片/链接 |
+| :--- | :--- | :--- | :--- |
+| [罗技 MX Master 3S 无线鼠标 人体工学](https://union-click.jd.com/jdc?e=0_2_0_NONE%7CMCP&p=JF8BAOEJK1olVQ4FVV5UC08XM28JGloWXwEFVl5ZCHtTXDdWRGtMGENDFlVDFhNSVzMXQA4KD1heSl5cCUoUAWgPGVsRXRlbEQIAOD1yATAAWTxPP2JaPyM5Ci8TVQtjbSsZUTYHVF9cCUMQAmgJK1sUXAQFVVdYCUMnM28JKw17XQcDVV9cCUgSCl8KGloXXQcHUFhUOEsQB2oBH1MRWg4CXF9tCEMTMzxYQw5RH19SCwgcUBQnM18LK2slXTYBZAAzCRgQBmcJEgl7AAMLBgEeSEJ5A2sMHlMWXAQDZFxcCUkVM18) | 约 529.0元 | 🖱️打工人必备！MX Master 3S静音顺滑，人体工学设计，手腕再也不酸了~ | [![图](http://t15.baidu.com/it/u=812867358,1743760388&fm=224&app=112&f=JPEG?w=500&h=500)](https://union-click.jd.com/jdc?e=0_2_0_NONE%7CMCP&p=JF8BAOEJK1olVQ4FVV5UC08XM28JGloWXwEFVl5ZCHtTXDdWRGtMGENDFlVDFhNSVzMXQA4KD1heSl5cCUoUAWgPGVsRXRlbEQIAOD1yATAAWTxPP2JaPyM5Ci8TVQtjbSsZUTYHVF9cCUMQAmgJK1sUXAQFVVdYCUMnM28JKw17XQcDVV9cCUgSCl8KGloXXQcHUFhUOEsQB2oBH1MRWg4CXF9tCEMTMzxYQw5RH19SCwgcUBQnM18LK2slXTYBZAAzCRgQBmcJEgl7AAMLBgEeSEJ5A2sMHlMWXAQDZFxcCUkVM18) |
+| [Anker 737 移动电源 240W 氮化镓](https://union-click.jd.com/jdc?e=0_2_0_NONE%7CMCP&p=JF8BARIJK1olXwICUl1fDU0VAl8IGlsVXwcCXFtYCEsUBV9MRANLAjZbERscSkAJHTdNTwcKBlMdBgABFksWA28KGlsdWAMCVF1bFxJSXzI4Wh9oLQJwNSw4bhtwUA0AcBxoFnhDNFJROE4XAm4JE1wUWgcyVF9cCkwWCmoJE2slXQcyFTBdCUkSCmkMHmsXXAcAVF9YDE0eM28OGF4WWwQDVlZbD0wnA2cMKwhFBVNGFgcNVx1WWzA4K2sWbTYyVG5eOBV5AjwOE14cDQNsCF9YWxRFVzFmHl8RXg8LUl1tCkoWAW04K2tIFARCBAYgaxB-ADBPGllCH3N3VTU-VCN5ATMBZjB9GA5KDyMCCwBBfmZxK14l) | 约 349.0元 | 🔋出差党狂喜！安克65W二合一充电宝，手机笔记本都能充，还能带上飞机~ | [![图](http://t13.baidu.com/it/u=2553962101,2686970068&fm=224&app=112&f=JPEG?w=500&h=500)](https://union-click.jd.com/jdc?e=0_2_0_NONE%7CMCP&p=JF8BARIJK1olXwICUl1fDU0VAl8IGlsVXwcCXFtYCEsUBV9MRANLAjZbERscSkAJHTdNTwcKBlMdBgABFksWA28KGlsdWAMCVF1bFxJSXzI4Wh9oLQJwNSw4bhtwUA0AcBxoFnhDNFJROE4XAm4JE1wUWgcyVF9cCkwWCmoJE2slXQcyFTBdCUkSCmkMHmsXXAcAVF9YDE0eM28OGF4WWwQDVlZbD0wnA2cMKwhFBVNGFgcNVx1WWzA4K2sWbTYyVG5eOBV5AjwOE14cDQNsCF9YWxRFVzFmHl8RXg8LUl1tCkoWAW04K2tIFARCBAYgaxB-ADBPGllCH3N3VTU-VCN5ATMBZjB9GA5KDyMCCwBBfmZxK14l) |
+| [绿联 100W 四口充电器 桌面充电站](https://union-click.jd.com/jdc?e=0_2_0_NONE%7CMCP&p=JF8BAQ0JK1olVQcAVF1YDEsSM28JGl0SWwQKVlxbAE8eMytXQwVKbV9HER8fA1UJWypcR0ROCBlQCgJDCEoWBWgOGVMXXwAKUFdCUQ5LXl9_Q1NBOH9JDj0dQA5wARxrRARWHFNEWFJtDUsWAm4AHFoSXDYCVV9fD0oeBm4AK2sVXDZDOl1dC00SCl8KGloXXQcHUFhUOEsQCmsIHV4WVAEBXFptCEMTMzxYQw5RH19SCwgcUBQnM18LK2slXTYBZAAzCRgRAD0PS117AF4GFVYIDB95A2YJE18RWQMyVl9cCkknM19AWDsUNnRyBBYZSx53YyZDUjlQBkRjVlkzCh1OfT1Xf1tTFXZmNiVbDR9xMw) | 约 599.0元 | 🔌桌面终结者！绿联100W氮化镓充电器，多设备同时快充，告别线材缠绕~ | [![图](http://t13.baidu.com/it/u=1745949773,1204442136&fm=224&app=112&f=JPEG?w=500&h=500)](https://union-click.jd.com/jdc?e=0_2_0_NONE%7CMCP&p=JF8BAQ0JK1olVQcAVF1YDEsSM28JGl0SWwQKVlxbAE8eMytXQwVKbV9HER8fA1UJWypcR0ROCBlQCgJDCEoWBWgOGVMXXwAKUFdCUQ5LXl9_Q1NBOH9JDj0dQA5wARxrRARWHFNEWFJtDUsWAm4AHFoSXDYCVV9fD0oeBm4AK2sVXDZDOl1dC00SCl8KGloXXQcHUFhUOEsQCmsIHV4WVAEBXFptCEMTMzxYQw5RH19SCwgcUBQnM18LK2slXTYBZAAzCRgRAD0PS117AF4GFVYIDB95A2YJE18RWQMyVl9cCkknM19AWDsUNnRyBBYZSx53YyZDUjlQBkRjVlkzCh1OfT1Xf1tTFXZmNiVbDR9xMw) |

Co-creation-projects/jack6249-GiftGeniusAgent/requirements.txt

@@ -0,0 +1,18 @@
+# HelloAgents框架
+hello-agents[all]>=0.1.0
+
+# LLM与搜索工具
+openai
+tavily-python
+numpy
+
+# Jupyter环境 & 异步修补 (新增 nest_asyncio)
+jupyter>=1.0.0
+notebook>=7.0.0
+nest_asyncio>=1.5.0
+
+# 环境变量管理
+python-dotenv>=1.0.0
+
+# MCP 协议支持 (新增 mcp)
+mcp>=0.1.0