赛事赛题生成器

技能文档

---
name: problem-set-generator
slug: problem-set-generator
displayName: 赛事赛题生成器
version: 1.1.1
category: ai-agent
display_name: 赛事赛题生成器
title: 赛事赛题生成器
author: 诺声(SynomosAI)
author_note: '"SynomosAI" / "SynomosAI" — 未申请实体注册、未申请商标注册（not a registered legal entity; no trademark registered）'
license: MIT
allowed-tools: Read, Glob, Grep, Bash, Write, WebFetch
license_scope: 代码（.py 文件）MIT；本 SKILL.md 与其中理论文本不在 MIT 覆盖范围内（见文末「License & attribution」）
description: 赛事赛题生成器——基于 UIBC 赛事规范的零依赖决策支持工具，JSON IR 输出，覆盖TH-EVT-002域。 适用场景：当需要赛事赛题生成器能力时使用本技能；触发词：赛事、判定器、TH-EVT-002、problem-set-generator。
tags: [赛事, 判定器, TH-EVT-002, problem-set-generator, 赛题生成, UIBC]
description_zh: "赛事赛题生成器——基于 UIBC 赛事规范的零依赖决策支持工具，JSON IR 输出，覆盖TH-EVT-002域。"
description_en: "Contest problem-set generator: zero-dependency decision support with JSON IR output (UIBC spec, domain TH-EVT-002)."
read_when:
  - 筹备 UIBC 竞赛要快速生成符合官方规范的题目
  - 工作流要求结构化输出而非自然语言
  - 需按 TH-EVT-002 领域标准产出赛题
  - 无外部脚本库/文件/代码执行环境的自动化场景
classification:
  skillhub: ["community-events"]
  iso_25010: ["Usability"]
risk_tier: medium
risk_rationale: "门禁批量补丁默认 medium（内部约定：本 skill 设计为中等风险）：调用 API/读写文件/本地执行，最小权限，密钥走 Secret；若实际涉对外代执行(push/发布/发送/上架)须人工复评，改 high 并走负责人批准。"
platforms: [windows, macos, linux]
platform_compatibility: [windows, macos, linux]
ai_protocol: true
spec: usk/1.0
interface:
  type: cli
  entry_point: problem-set-generator.py
  runtime: python3
  call_pattern: args
input_schema:
  theme:
    type: string
    description: "赛题主题(默认 迁移的判定与问责)"
    required: false
  difficulty:
    type: string
    description: "难度: easy/mid/hard"
    required: false
  count:
    type: string
    description: "生成数量 1-10"
    required: false
  demo:
    type: bool
    description: "跑内置冒烟案例"
    required: false
  json:
    type: bool
    description: "输出 JSON IR（默认）"
    required: false
output_schema:
  stdout: 结构化 JSON IR（含 result / rc / error_code / warnings）
  exit_code: rc=0 正常 / rc=1 告警或从严推定 / rc=2 输入不足或非法
capabilities:
  - education
  - problem-set
  - generation
permissions:
  network: false
  filesystem: false
  subprocess: false
requirements:
  python_packages: []
  min_python: "3.8"
changelog: "1.1.1: Bilingual rewrite (full EN+ZH, English first); 1.0.0: initial upload (USK v3)"
---

> 凡自治之物 · LGD 治理学 · 创造术家族 ｜ 评测协议：skill-quality-gate

## English

### When to use
1. While preparing UIBC-related contests, when you need to quickly produce problem content that complies with official contest rules.
2. When the workflow requires structured data output rather than natural-language description, and must strictly follow the TH-EVT-002 domain standard.
3. In automated decision-support scenarios where external script libraries, files or code-execution environments are unavailable.

### Usage
Pass parameter configurations directly through the skill interface; no local scripts or config files need to be loaded. Core inputs: `domain` specifying the TH-EVT-002 domain, `difficulty_level` setting the difficulty tier, and `output_format` forcing JSON IR return. The skill directly generates format-compliant JSON blocks without extra file operations or script parsing.

### Workflow
1. Confirm the current task context matches UIBC generation needs; define the target problem domain and count.
2. Initialize input variables per the parameter contract; ensure `output_format` points to JSON IR.
3. Invoke the skill and emphasize TH-EVT-002 compliance constraints in the prompt.
4. Receive the returned raw data and verify its structure against the preset IR schema.
5. Feed the generated JSON into downstream workflows, or perform field extraction and completeness checks.
6. Deliver the problem set; ensure the whole run is zero-dependency and spec-compliant.

### Evidence & safety statement
- Evidence: TBD (no evaluation / badge link yet; update here once produced)
- Version: 1.1.1
- Safety: no independent security scan has been performed (TBD); the conclusion is pending and no endorsement is given until it is completed.

## 中文

### 何时用
1. 在筹备 UIBC 相关竞赛时，需要快速生成符合赛事官方规范的题目内容。
2. 当工作流要求输出结构化数据而非自然语言描述，且需严格遵循 TH-EVT-002 领域标准时。
3. 在进行自动化决策支持场景下，当前无法依赖外部脚本库、文件或代码执行环境时。

### 用法
通过调用该技能接口直接传入参数配置，无需加载本地脚本或配置文件。核心输入包含：`domain` 指定 TH-EVT-002 相关领域，设置 `difficulty_level` 定义难度等级，配置 `output_format` 强制返回 JSON IR。技能将直接生成符合格式的 JSON 数据块，无需用户进行额外的文件操作或脚本解析。

### 流程
1. 确认当前任务上下文符合 UIBC 赛事生成需求，明确目标赛题领域与数量。
2. 根据参数要求初始化输入变量，确保 `output_format` 指向 JSON IR 格式。
3. 发送指令调用技能，并在提示词中强调 TH-EVT-002 领域的合规性约束。
4. 接收系统返回的原始数据，验证其结构是否符合预设的 IR Schema 标准。
5. 将生成的 JSON 数据导入后续工作流，或进行必要的字段提取与完整性检查。
6. 完成问题集交付，确保整个过程为零依赖运行且符合规范输出要求。

### 证据与安全声明

- 证据：TBD（暂无评测/徽章链接；产生后在此更新实际链接）
- 版本：1.1.1
- 安全：未做独立安全扫描（TBD），结论待补；补全前不作推荐背书。

使用说明

# problem-set-generator · 赛事赛题生成器 / Contest Problem-Set Generator

[![LGD](https://img.shields.io/badge/LGD-Governance-blue)](https://github.com/zhaoxinghua09-cell/lgd-theory)

## English

> AI-assisted decision-support CLI for UIBC contest operations. NOT final adjudication — verify with the organizing committee. Text verified: 2026-09-26.

### What it does
Generates **structured problem-set drafts** (background / task / constraints / deliverables / scoring) per **UIBC problem rules + season theme "Migration decisions and accountability"** (rule source pending manual verification; the latest UIBC charter prevails). Every problem is marked `review_status: pending` — committee review is required before publishing.

- **Strict on bad input**: invalid `difficulty` (not easy/mid/hard) or invalid/out-of-range `count` → `rc=2` — no silent fallback to mid/3. Custom (non-season) theme → adaptation warning, `rc=1` — never silently passed as season-compliant.
- **Honest boundary**: drafts are 3-scenario templates differentiated by difficulty-requirement sections; the committee must refine/dedupe before publishing (noted in output `notes`).

### Install & run (zero-dependency, Python 3.8+)
```bash
python problem-set-generator.py --difficulty hard --count 1                    # season theme → rc=0
python problem-set-generator.py --theme 跨境数据判定 --difficulty easy --count 1   # custom theme → warning, rc=1
python problem-set-generator.py --difficulty banana                            # invalid → rc=2
python problem-set-generator.py --count 11                                     # out of range → rc=2
python problem-set-generator.py --demo                                         # 3 built-in smoke cases
```
Params: `theme` (default = current season theme) / `difficulty` (easy/mid/hard) / `count` (1-10).

### Output
Deterministic JSON IR with `rc` (**0** = ok / **1** = completed with warnings / **2** = invalid input), `problems[]` (each with `spec`, `review_status`, `ai_note`), `warnings`, and `aigc_mark` (GB 45438-2025 metadata; `is_generated: true`, draft text is AI-generated).

Real excerpt (`--demo`, case 3; `spec` folded; other values verbatim):
```json
{
  "input": {"theme": "跨境数据判定", "difficulty": "easy", "count": "1"},
  "result": {
    "problems": [{"index": 1, "difficulty": "入门", "review_status": "待审（组委会人工审核后方可发布）"}],
    "warnings": ["自定义主题「跨境数据判定」非当前赛季主题「迁移的判定与问责」：是否采用须组委会确认赛题适配性。"]
  },
  "rc": 1
}
```

### Rule-source checklist pending manual verification
| Rule | Use | Status |
|---|---|---|
| UIBC problem rules | Problem structure & difficulty tiers | Pending |
| Season theme "迁移的判定与问责" | Default theme & custom-theme warning basis | Pending |

### Compliance & IP
- LGD-Powered badge · theory TH-EVT-002 · ambassador Nuosheng @ contest chief voice
- This tool is decision support; human review is required.
- **Layered licence**: code `problem-set-generator.py` = MIT; this `README.md` / `SKILL.md` and all theory text are **not covered by MIT** (all rights reserved).

### Family set
| Contest stage | Tool | Ambassador |
|---|---|---|
| Registration / submission portal | contest-submission-portal | Nuosheng |
| Problem generation | problem-set-generator | Nuosheng |
| Judging rubric scoring | judging-rubric-grader | Nvlv |
| Anti-cheat / identity verification | anti-cheat-identity | Nuowei |
| Submission plagiarism check | originality-check | Nuoyuan |

## 中文

> AI 辅助决策支持 CLI，服务 UIBC 赛事运营。**非最终裁决**——须以组委会复核为准。文本核验日期：2026-09-26。

### 它做什么
按 **UIBC 赛题规范 + 赛季主题「迁移的判定与问责」**（规则源待人工核对；以最新 UIBC 章程为准）生成**结构化赛题草稿**（背景/任务/约束/交付物/评分）。每题标记 `review_status: 待审`——发布前须组委会审核。

- **坏输入从严**：`difficulty` 非法（非 easy/mid/hard）或 `count` 越界 → `rc=2`——不静默回落到 mid/3。自定义（非赛季）主题 → 适配告警 `rc=1`——绝不静默当作赛季合规通过。
- **边界诚实**：草稿为 3 场景模板、按难度-要求段区分；发布前组委会须精修/去重（输出 `notes` 已注明）。

### 安装与运行（零依赖，Python 3.8+）
```bash
python problem-set-generator.py --difficulty hard --count 1                    # 赛季主题 → rc=0
python problem-set-generator.py --theme 跨境数据判定 --difficulty easy --count 1   # 自定义主题 → 告警，rc=1
python problem-set-generator.py --difficulty banana                            # 非法 → rc=2
python problem-set-generator.py --count 11                                     # 越界 → rc=2
python problem-set-generator.py --demo                                         # 3 个内置冒烟案例
```
参数：`theme`（默认=当前赛季主题）/ `difficulty`（easy/mid/hard）/ `count`（1-10）。

### 输出
确定性 JSON IR，含 `rc`（**0**=正常 / **1**=带告警完成 / **2**=输入非法）、`problems[]`（每条含 `spec`、`review_status`、`ai_note`）、`warnings` 与 `aigc_mark`（GB 45438-2025 元数据；`is_generated: true`，草稿文本为 AI 生成）。

### 规则源待人工核对清单
| 规则 | 用途 | 状态 |
|---|---|---|
| UIBC 赛题规范 | 赛题结构与难度分级 | 待人工核对 |
| 赛季主题「迁移的判定与问责」 | 默认主题与自定义主题告警依据 | 待人工核对 |

### 合规与知识产权
- LGD-Powered 徽章 · 理论 TH-EVT-002 · 大使诺声 @ 赛事首席发声人
- 本工具为决策支持，须人工复核。
- **分层许可**：代码 `problem-set-generator.py` = MIT；本 `README.md` / `SKILL.md` 及一切理论文本**不在 MIT 覆盖范围**（保留所有权利）。

### 家族合集
| 赛事环节 | 工具 | 大使 |
|---|---|---|
| 报名/提交门户 | contest-submission-portal | 诺声 |
| 赛题生成 | problem-set-generator | 诺声 |
| 评审打分 | judging-rubric-grader | 诺律 |
| 反作弊/身份核验 | anti-cheat-identity | 诺卫 |
| 作品查重 | originality-check | 诺源 |

---

## License & attribution
**分层许可（务必并读；代码许可不覆盖文档与理论文本）**

- **代码**（本包 `.py` 文件）：**MIT**
- **本 `README.md` / `SKILL.md` 及其中的理论文本、方法论与一切理论表述**：**不在 MIT 覆盖范围内**，保留所有权利（All rights reserved）

```
© 2026 SynomosAI. All rights reserved.
理论署名 (attribution) : LGD（Lifecycle Governance Doctrine / 全程治理论）— SynomosAI initiative
名称状态 (name status)  : "SynomosAI" — 未申请实体注册、未申请商标注册
                        (not a registered legal entity; no trademark registered)
生产参考部署 (production reference, self-reported) : 无（未声明外部生产参考部署）
                    ← 非认证、非背书、非监管认可（not a certification or endorsement）
代码许可 (code license) : 本包未附 LICENSE 文件（许可待定）
                    本文本与理论表述不在任何代码许可覆盖范围内
引用格式 (cite as)      : 本资产无 DOI
```
（上列为《LGD 对外表述规范》§3.2 统一块，**整体复制、未删改**；发布/重提前须照最新版本复核。）

如何安装此技能？

访问技能市场，点击「安装」按钮，按提示将技能包放入 AI 编程助手的 skills 目录即可。