← Back to Skills Marketplace
jhl-det

Tencent Yuanbao Gaokao - Score-to-Rank Lookup

by jhl-Det · GitHub ↗ · v0.1.2 · MIT-0
cross-platform ✓ Security Clean
91
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install score-range
Description
An information retrieval assistant for GaoKao score-by-score rankings, helping candidates query their provincial ranking based on their score, estimate the c...
README (SKILL.md)

核心原则

根据用户的问题,依次提取以下五类信息: 地区(place)、年份(year)、选科(classify)、分数(score)、省排名(rank)


Step 1:提取地区(place)

读取 ./ref/place.json 中的地区列表,将用户 query 中提及的地区与列表进行匹配。

匹配规则:

  • 支持别称/简称(如"魔都"→ 上海、"首都"→ 北京、"闽南"→ 福建、"桂"→ 广西)
  • 支持容错匹配(如存在错别字或非标准表述,尽量推断正确地区)
  • 支持城市到省份的映射(如"福州"->"福建"、"深圳"->"广东")
  • 如果用户提及了多个省份,则每个省份之间使用空格隔开,如"北京 上海 广东"
  • 如果用户想查询所有省份(如"各省"、"各地区"),则返回 "place" : "全省"
  • 以列表中的标准名称作为最终输出值

若无法从 用户query 中识别地区,则询问用户:"您想查询哪个省份的信息呢?"。


Step 2:提取年份(year)

  • 若用户明确提及年份(具体的数字),直接提取年份
  • 若用户提及的是模糊的年份,诸如今年、去年、近三年的说法,今年指2026,去年指2025
  • 若用户未提及年份,则不返回 "year" 字段
  • 年份必须是四位数字,如 "25年"->"2025"
  • 若用户提及了多个年份,则每个年份之间使用空格隔开,如"2025 2026"

Step 3:提取选科(classify)

读取 ./ref/classify.json 中的选科列表,将用户 query 中提及的选科与列表进行匹配。

核心难点:新老高考政策差异 中国各省份在不同年份实行的高考政策不同(分为:传统文理科、3+3新高考、3+1+2新高考),选科名称会随之变化。你必须结合 Step 1 的【地区】和 Step 2 的【年份】进行动态匹配与转换。

下表列出各省份切换新高考的年份,切换年份之前为传统文理科模式

模式 省份 启用年份
3+3 上海、浙江 2017
3+3 北京、天津、山东、海南 2020
3+1+2 河北、辽宁、江苏、福建、湖北、湖南、广东、重庆 2021
3+1+2 甘肃、吉林、黑龙江、安徽、江西、贵州、广西 2024
3+1+2 山西、内蒙古、河南、四川、云南、陕西、青海、宁夏 2025

内蒙古蒙授生(蒙授文科/蒙授理科)不受以上规则约束,始终保持独立分类。

匹配与自动纠错规则:

  1. 传统高考模式
    • 标准选科为:文科 理科
    • 映射:用户提及"文史类"→ 文科;"理工类"→ 理科
  2. 3+3 新高考模式
    • 标准选科为:综合(不分文理)
    • 纠错:若用户提及"文科/理科/物理/历史",自动映射为:综合
  3. 3+1+2 新高考模式
    • 标准选科为:物理 历史
    • 纠错:若用户提及"理科/理工",自动映射为:物理;提及"文科/文史",自动映射为:历史

若用户未提及选科,则不返回 "classify" 字段


Step 4:提取分数(score)

若用户明确说分数(如"考了 600 分"、"分数是 580"、"600 分能排第几"、"我考了 600"),则返回 "score" 字段。

  • score 必须是数字,如"考了 600 分"->"600"
  • 若无法提取到数字信息,则禁止 "score" 字段

Step 5:提取省排名(rank)

若用户明确说位次/排名(如"位次 1000"、" 400 名"、"省排 500 名"、"排名第 2000"、"我排在第 1000 位"),则返回 "rank" 字段。

  • 注意中文数字需转成整数,如"5万名"->"50000"
  • rank 必须是数字,如"排名第 1000 位"->"1000"
  • 若无法提取到数字信息,则禁止 "rank" 字段

Step 6:调用脚本查询数据(db_info)

确认地区、年份、选科后,调用以下脚本获取真实数据: 注意,若用户未提及分数或排名,也必须调用脚本获取数据。 禁止直接访问API获取完整数据,必须通过脚本获取数据。 脚本路径: ./fetch_data.py 调用方式:

python3 ./fetch_data.py --place {place...} --year {year...} --classify {classify...} --score {score...} --rank {rank...}

Step 7:根据脚本查询到的数据判断是否符合考生的意图

  • 根据“选科”信息,保留符合query意图的数据。

Step 8:总结

结合上述内容,对用户的问题进行回答。 注意需要简明扼要。

Step 9:反问排名/分数【非必要】

如果用户没有提及分数/排名,需要结合用户已有的信息(地区(place)、年份(year)、选科(classify))反问用户分数/排名。如“请问您的分数或者省排名是多少呢?我来帮您查询对应的分数/排名~” 然后再使用Step 4~8的步骤进行回答。

Usage Guidance
This skill looks safe for its stated GaoKao score/rank lookup purpose. Before installing, be aware that it executes a local Python script and queries Tencent's GaoKao data endpoint with province/year/subject information; avoid adding unnecessary personal details beyond what is needed for the lookup.
Capability Analysis
Type: OpenClaw Skill Name: score-range Version: 0.1.2 The skill bundle is a legitimate tool for querying Chinese GaoKao (college entrance exam) score rankings. It extracts user parameters such as region, year, and score, then fetches data from a Tencent-owned domain (gaokao.search.qq.com) using the fetch_data.py script. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found; a minor parameter naming inconsistency ('--ip') in the script appears to be a non-functional copy-paste error.
Capability Assessment
Purpose & Capability
The capability matches the stated purpose of looking up GaoKao score/rank tables, with a narrow set of extracted fields: province, year, subject category, score, and rank.
Instruction Scope
The instructions are task-specific and do not attempt to override user intent, create hidden authority, or make unrelated content authoritative.
Install Mechanism
There is no install spec and no dependencies beyond an included Python script using the standard library, but the metadata does not declare the python3 runtime that SKILL.md expects.
Credentials
The helper makes a purpose-aligned external request to gaokao.search.qq.com using query context such as province, year, and subject category; it does not request credentials or broad local files.
Persistence & Privilege
The artifacts show no persistent background behavior, privilege escalation, credential use, file writes, or account mutation.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install score-range
  3. After installation, invoke the skill by name or use /score-range
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.2
update display name
v0.1.1
No functional or file changes in this version. - No code or documentation files were modified. - SKILL.md description updated from Chinese to English for the main description line; all usage instructions and extraction rules remain unchanged. - No new features or bug fixes included.
v0.1.0
- Initial release of "score-range" skill. - Enables users to retrieve provincial ranking by score, estimate score ranges by rank, or access segmented score tables for the 高考. - Automatically extracts region, year, subject selection, score, and rank from queries. - Handles province/city naming variations, typos, and subject classification rules based on region and exam year. - Dynamically matches old/new 高考 policies for subject extraction. - Interactively asks for missing key information (region, score, or rank) as needed.
Metadata
Slug score-range
Version 0.1.2
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is Tencent Yuanbao Gaokao - Score-to-Rank Lookup?

An information retrieval assistant for GaoKao score-by-score rankings, helping candidates query their provincial ranking based on their score, estimate the c... It is an AI Agent Skill for Claude Code / OpenClaw, with 91 downloads so far.

How do I install Tencent Yuanbao Gaokao - Score-to-Rank Lookup?

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

Is Tencent Yuanbao Gaokao - Score-to-Rank Lookup free?

Yes, Tencent Yuanbao Gaokao - Score-to-Rank Lookup is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Tencent Yuanbao Gaokao - Score-to-Rank Lookup support?

Tencent Yuanbao Gaokao - Score-to-Rank Lookup is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Tencent Yuanbao Gaokao - Score-to-Rank Lookup?

It is built and maintained by jhl-Det (@jhl-det); the current version is v0.1.2.

💬 Comments