← Back to Skills Marketplace
deusyu

car-advisor

by Rainman · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
282
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install car-advisor
Description
实时汽车问答与对比分析系统。当用户询问任何买车、选车、汽车参数对比、车型评测、价格分析相关问题时触发此 Skill。 触发场景(只要涉及以下任一情形就必须使用此 Skill): - 车型参数对比:"小米SU7和Model 3哪个好"、"国产车和特斯拉对比" - 配置/价格查询:"Model Y 焕新版座椅加热有吗...
README (SKILL.md)

Car Advisor Skill

一个基于实时数据的汽车问答与对比系统。核心原则:不用训练数据猜,用实时检索说话


工作流程总览

用户提问
   ↓
[Step 1] 解析意图 → 确定车型、对比维度、问题类型
   ↓
[Step 2] 并行数据采集 → 参数 + 价格 + 真实评论
   ↓
[Step 3] 数据整理 → 生成结构化表格或分析报告
   ↓
[Step 4] 输出结论 → 附数据来源、注明采集时间

Step 1:意图解析

收到用户问题后,先在内部识别:

维度 提取内容
车型列表 涉及哪些具体车型(品牌+车系+年款)
问题类型 参数查询 / 横向对比 / 口碑评价 / 购车建议 / 价格分析
核心关注点 用户最在意的维度(续航/智驾/内饰/价格/空间…)
数据新鲜度要求 是否需要最新价格/最新配置(如焕新版、新款)

如果车型信息模糊,可以先用 web_search 确认当前在售的具体版本(例如"Model Y 2025款"的正式名称)。


Step 2:数据采集策略

2.1 数据源优先级

详细数据源列表见 references/data-sources.md

优先级1 — 官方页面(最权威)

  • 特斯拉:https://www.tesla.cn/modelyhttps://www.tesla.cn/model3
  • 小米:https://www.xiaomiev.com
  • 问界/AITO:https://www.aito.com
  • 理想:https://www.lixiang.com
  • 比亚迪/各国产品牌官网

优先级2 — 垂直媒体参数库(最全面)

  • 懂车帝参数页:https://www.dongchedi.com/auto/params-carIds-x-{car_id}
  • 汽车之家参数对比:https://www.autohome.com.cn/config/
  • web_search 搜索"车型名 参数配置 懂车帝"快速定位页面

优先级3 — 真实评论/口碑

  • 懂车帝口碑:搜索 {车型名} 口碑 懂车帝
  • 汽车之家车主点评:搜索 {车型名} 车主评价 汽车之家
  • 知乎用车报告:搜索 {车型名} 用车体验 知乎
  • 媒体评测:搜索 {车型名} 深度评测 2024

优先级4 — 价格/销量数据

  • 当前售价:搜索 {车型名} 最新价格 2025(价格变动频繁,必须实时)
  • 销量数据:搜索 {车型名} 销量 最新

2.2 并行检索策略

对于多车型对比,并行发起多个搜索,不要串行等待。典型搜索组合:

同时搜索:
- "{车型A} 配置参数 {年款}"
- "{车型B} 配置参数 {年款}"
- "{车型A} vs {车型B} 对比 评测"
- "{车型A} 车主评价 缺点"
- "{车型B} 车主评价 缺点"

Step 3:数据整理规范

3.1 参数对比表模板

生成对比表时,维度根据用户关注点动态调整,但建议包含以下核心维度:

基础信息层

维度 说明
车型/版本 品牌+车系+具体版本+年款
官方起售价 注明是否含补贴,标注采集日期
车身尺寸 长×宽×高,轴距
定位 级别(A/B/C级)、车身形式

动力/续航层(新能源重点)

维度 说明
电池容量 kWh
CLTC续航 公里
最大功率 kW/马力
百公里加速
充电功率 kW(快充峰值)
能耗 kWh/100km

内饰/舒适层

维度 说明
座椅配置 加热/通风/按摩/电动调节,注明前后排
中控屏 尺寸+系统
后排屏 有无,尺寸
车机系统 品牌/版本,手机互联能力
氛围灯 有无,分区数
音响 品牌+扬声器数量
空气悬架 有无
特色功能 香氛/冰箱/V2L/露营模式

智驾层

维度 说明
智驾方案 品牌/版本
激光雷达 有无
高速NOA 是否支持及收费情况
城市NOA 是否支持及覆盖城市数
智驾硬件 算力 TOPS

安全/空间

维度 说明
安全气囊数
后备箱容积 L(含前备箱)
纯平放倒 是否支持,长度

3.2 评论整合规范

在对比表后,附上真实评论摘要,格式如下:

## 车主真实反馈

### [车型A]
**好评高频词**:[提炼3-5个]
**投诉集中点**:[提炼3-5个]
**代表性车主评论**:
- "..." (来源:懂车帝/汽车之家,评分X星)
- "..."

### [车型B]
...

> 数据来源:懂车帝口碑、汽车之家车主点评
> 采集时间:{当前日期}
> 注:评论为用户主观表达,仅供参考

Step 4:输出规范

4.1 对比结论结构

## 一句话总结
[直接给出核心结论,不废话]

## 参数对比表
[完整表格]

## 各维度胜出分析
- 续航/能效:XX胜出,原因...
- 智驾:XX胜出,原因...
- 内饰/舒适:XX胜出,原因...
- 性价比:XX胜出,原因...

## 车主真实反馈
[评论摘要]

## 购买建议
如果你更在意[X],选[车型A]
如果你更在意[Y],选[车型B]

> 数据采集时间:{日期} | 价格以官网最新为准

4.2 单车型查询结构

## [车型名] 当前配置一览

[关键参数,重点回答用户问题]

## 各版本差异
[如有多版本,列出关键差异]

## 真实车主评价
[评论摘要]

## 综合评价
[一段客观评述]

特殊场景处理

购车预算推荐

当用户给出预算范围时:

  1. 先搜索该预算区间当前热门车型(搜索"20-30万 新能源SUV 2025 推荐")
  2. 筛选出3-5款候选,做简要对比
  3. 按用户明确提及的偏好(家用/运动/长途/城市)给出排序建议

价格变动查询

价格是高度动态的信息(特斯拉尤其频繁调价):

  • 必须实时搜索,不得用训练数据中的价格
  • 搜索时加入"最新""2025"等时间词确保信息新鲜度
  • 如搜索结果显示价格有争议,注明"以官网实时为准"

无法获取数据时

如果某个具体参数无法通过搜索确认:

  • 明确说明"未能从实时数据中确认此参数"
  • 提供参考来源链接让用户自行核查
  • 不用训练数据填充(可能过时)

数据质量原则

  1. 时效性优先:搜索时始终加年份词("2025"或"最新"),确保信息非过时版本
  2. 交叉验证:重要参数(价格、续航)尽量从2个来源确认
  3. 来源透明:每条关键数据标注来源(官网/懂车帝/汽车之家)
  4. 主观与客观分离:参数表用客观数据,评论区用主观反馈,两者不混用
  5. 不确定性诚实:数据有争议时说明,不强行给出确定答案

参考资源

  • 详细数据源列表见 references/data-sources.md
  • 对比维度优先级矩阵见 references/compare-dimensions.md
Usage Guidance
This skill appears coherent and does what it says: perform real-time web searches and assemble car comparisons. Before installing, consider: (1) privacy — searches and user queries will go to whatever web_search provider the agent uses (they may be logged by that provider and by the target websites); (2) volume and rate limits — the skill advocates parallel searches and may perform many queries for multi-vehicle comparisons (this can hit rate limits or trigger blocking on some sites); (3) accuracy — the skill relies on scraped/third-party pages that can be out-of-date or inconsistent, so verify critical data (price, subscription requirements) against official brand pages; (4) optional paid APIs — if you later supply API keys for juhe/wapi/dongchedi, grant only the minimal permissions and store keys in secure env vars; (5) autonomous invocation — if you are concerned about automatic network activity, consider restricting autonomous execution or adding usage limits. Overall there are no obvious coherence or credential-exfiltration issues.
Capability Analysis
Type: OpenClaw Skill Name: car-advisor Version: 1.0.0 The skill bundle provides structured instructions and reference materials for a car advisory system. It directs the AI agent to use real-time web searches on legitimate automotive platforms (e.g., Dongchedi, Autohome, and official manufacturer websites) to retrieve vehicle specifications, pricing, and user reviews. No malicious code, data exfiltration attempts, or harmful prompt injections were found; the logic is entirely consistent with the stated purpose of automotive comparison and analysis.
Capability Assessment
Purpose & Capability
Name/description (real-time car Q&A and comparisons) match the SKILL.md: it describes intent parsing, web searches of official sites and auto portals, structured comparison output, and cites appropriate sources. No unrelated environment variables, binaries, or config paths are requested.
Instruction Scope
Instructions require broad, parallel web_searches and scraping-like retrieval from many public automotive sites (official pages, dongchedi, autohome, media, forums). This is coherent for the purpose, but the policy to 'always trigger when a topic involves specific models' may cause frequent external queries (privacy/rate-limit considerations). The SKILL.md does not direct reading of local files or other unrelated system state.
Install Mechanism
No install specification or code files — instruction-only skill. This is the lowest-risk install model and consistent with the described behaviour.
Credentials
No environment variables or credentials are required. The references mention third-party paid APIs (wapi.cn, juhe.cn, dongchedi enterprise keys) that would require keys if used, but the skill does not demand those credentials in its manifest — provisioning them would be optional and should be done deliberately.
Persistence & Privilege
Skill is not always-included (always: false) and does not request to modify agent/system configs or persistent credentials. Autonomous invocation is allowed (platform default) but not combined with other privilege escalations.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install car-advisor
  3. After installation, invoke the skill by name or use /car-advisor
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release
Metadata
Slug car-advisor
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is car-advisor?

实时汽车问答与对比分析系统。当用户询问任何买车、选车、汽车参数对比、车型评测、价格分析相关问题时触发此 Skill。 触发场景(只要涉及以下任一情形就必须使用此 Skill): - 车型参数对比:"小米SU7和Model 3哪个好"、"国产车和特斯拉对比" - 配置/价格查询:"Model Y 焕新版座椅加热有吗... It is an AI Agent Skill for Claude Code / OpenClaw, with 282 downloads so far.

How do I install car-advisor?

Run "/install car-advisor" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is car-advisor free?

Yes, car-advisor is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does car-advisor support?

car-advisor is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created car-advisor?

It is built and maintained by Rainman (@deusyu); the current version is v1.0.0.

💬 Comments