Shike Gift Advisor

AI礼物推荐专家 - 根据收礼人关系、场合、预算智能推荐礼物。支持禁忌过滤、个性化祝福语。Use when: 礼物推荐, 生日礼物, 节日礼物, gift recommendation.

Security Scan

Instruction-only gift-recommendation skill; its declared inputs and behavior are coherent with its description and it does not request credentials or install code, though some implementation details (knowledge base, purchase links) are unspecified.

Purpose & Capability

Name and description match the runtime instructions: recommending gifts, filtering cultural taboos, and generating personalized greetings. The skill declares no binaries, env vars, or installs, which is proportionate for an advice/recommendation assistant.

Instruction Scope

SKILL.md stays within the stated purpose and does not instruct the agent to read local files, access environment variables, or send data to external endpoints. One ambiguity: it references calling a 'knowledge base' and providing 'purchase suggestions' but does not define a data source or external API — this is an implementation gap (not a direct security contradiction) and could affect privacy depending on how the agent resolves those steps.

Install Mechanism

No install spec and no code files to execute. Instruction-only skills are lowest-risk from an installation perspective.

Credentials

The skill requires no environment variables, credentials, or config paths. There are no disproportionate secret requests.

Persistence & Privilege

always is false and the skill does not request persistent system-level privileges. Autonomous invocation is allowed (platform default) but is not combined with other red flags.

Assessment

This skill appears internally consistent and does not request credentials or install code. Before installing or using it, consider: (1) It mentions a 'knowledge base' and 'purchase suggestions' but gives no implementation details — ask the provider where recommendations/links come from (local model knowledge vs external APIs). (2) Do not submit sensitive personal data (bank details, identity numbers, medical records); the skill may ask about preferences and taboos, which is fine, but avoid oversharing. (3) The SKILL.md lists an email and WeChat for paid services—treat external contact channels as separate third parties and be cautious about sending private information over them. (4) Performance claims (>99% accuracy, <1s response) are optimistic; test the skill with sample queries to evaluate quality. If you need stronger privacy guarantees or external integrations (e.g., shopping links), request explicit details about data flow and external endpoints from the publisher.

Gift Advisor - AI礼物推荐专家

送礼不是交易，是情感的传递。选礼物，就是选心意。

---

核心理念

关系识别 → 场合分析 → 预算匹配 → 禁忌过滤 → 个性化推荐

---

核心功能

1. 智能推荐

• 自然语言解析需求
• 3-5个候选礼物推荐
• 每个礼物包含：名称、价格、理由、适配度

2. 禁忌过滤

• 20+文化禁忌（钟、鞋、梨、伞等）
• 场合禁忌匹配
• 个人忌讳验证
• 提供安全替代品

3. 个性化祝福

• 多风格选择（标准、深情、幽默、简洁、文艺）
• 方言支持（粤语、四川话、上海话、东北话）
• 根据关系和礼物定制

---

适用场景

关系支持

• 家人：妈妈、爸爸、长辈、孩子
• 情侣：女友、男友、老公、老婆
• 朋友：闺蜜、兄弟、同事
• 商务：领导、客户、老师

场合支持

• 生日、节日、纪念日
• 情人节、母亲节、父亲节
• 春节、中秋、圣诞节
• 乔迁、结婚、毕业、升职
• 道歉、感谢、表白

---

使用方式

自然语言输入

用户：送妈妈的生日礼物，预算500元左右

输出示例

{
  "recommendations": [
    {
      "name": "护肤品套装",
      "price": 699,
      "reason": "妈妈日常保养需要，实用贴心",
      "score": 95
    },
    {
      "name": "丝巾",
      "price": 299,
      "reason": "优雅大方，适合各种场合",
      "score": 88
    }
  ],
  "blessing": "生日快乐！愿您年年有今日，岁岁有今朝！妈妈，您辛苦了！",
  "taboo_report": {
    "summary": "✅ 无禁忌礼物"
  }
}

---

工作流程

1. 信息收集
   - 解析收礼人关系
   - 识别场合类型
   - 提取预算范围
   - 了解偏好特征

2. 候选筛选
   - 调用知识库匹配
   - 过滤禁忌礼物
   - 计算适配分数
   - 排序推荐列表

3. 禁忌检查
   - 文化禁忌扫描
   - 场合禁忌匹配
   - 个人禁忌验证
   - 生成禁忌报告

4. 祝福语生成
   - 选择祝福模板
   - 个性化调整
   - 情感适配

5. 结果呈现
   - 格式化输出
   - 展示理由
   - 提供购买建议

---

关键检查点

检查点1：需求完整性

触发时机：信息收集后 用户确认：收礼人关系、场合、预算是否明确

检查点2：禁忌确认

触发时机：禁忌检查后 用户确认：是否有特殊文化禁忌或个人忌讳

检查点3：推荐验收

触发时机：呈现结果前 用户确认：推荐是否符合预期，是否需要调整

---

禁忌库示例

礼物	禁忌原因	替代方案
钟	谐音"送终"	电子闹钟、沙漏
鞋	谐音"邪"	舒适的拖鞋（避开正装鞋）
梨	谐音"离"	苹果、橙子
伞	谐音"散"	防晒帽、雨衣
蜡烛	祭祀寓意	香薰蜡烛、LED灯

---

性能指标

• 禁忌过滤准确率：>99%
• 祝福语个性化满意度：>85%
• 响应时间：<1秒
• 推荐适配度：>90%

---

技术支持

免费版：当前版本（MIT-0 license）

付费服务：

• 定制礼物库：¥500-1000
• 企业礼品方案：¥2000-5000
• 技术咨询：¥300/小时

联系方式：

• Email：[e-mail]
• 微信：ailvyou88999

---

支持场景（完整列表）

关系

妈妈、爸爸、女友、男友、老婆、老公、朋友、闺蜜、兄弟、长辈、孩子、老师、领导、同事、客户

场合

生日、情人节、纪念日、母亲节、父亲节、教师节、春节、中秋、乔迁、结婚、道歉、感谢、表白、毕业、升职、退休、看望病人