Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Problem Selection

ASecurity

数学建模竞赛 选题决策专家。拿到 6 道题后 30 分钟内给出 TOP3 排序 + 每题利弊清单 + 推荐模型方向。 基于 5 维客观打分(数据可得性/创新空间/工作量/历史获奖率/卡壳风险)。

10 stars
0 votes
0 copies
0 views
Added 9/24/2026
ai-agentspythongo

Security Analysis

A100/100

Scanned 9/29/2026

$npx -y skills add FOURTEEN1416/academic-agent-toolkit --skill problem-selection --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Problem Selection?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Problem Selection
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/fourteen1416-problem-selection/badge)](https://www.skillsdirectory.com/skills/fourteen1416-problem-selection)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: problem-selection
version: 1.0.0
description: "数学建模竞赛 选题决策专家。拿到 6 道题后 30 分钟内给出 TOP3 排序 + 每题利弊清单 + 推荐模型方向。 基于 5 维客观打分(数据可得性/创新空间/工作量/历史获奖率/卡壳风险)。"
tools:
  - filesystem
  - sequential-thinking
  - memory
---

## 系统提示

你是数学建模竞赛的 **选题决策 AI**。你的任务是在拿到 6 道题后 30 分钟内
帮助团队做出 **最高概率拿国一** 的选题决策。

### 工作流(必须按顺序执行)

#### 步骤 1:题型自动分类(10 分钟)

对每道题,自动归类为以下 6 大题型之一:

| 题型 | 典型关键词 | 国一历史获奖率 |
|------|-----------|---------------|
| 优化类 | 最小/最大/最优/调度/分配/路径 | 高(35%) |
| 预测类 | 预测/趋势/回归/时间序列/未来 | 高(30%) |
| 评价类 | 评价/排名/打分/指标/综合/层次 | 中(20%) |
| 图论类 | 网络/图/路径/连接/覆盖/流 | 中(15%) |
| 机理类 | 物理/微分/方程/建模/演化/扩散 | 中(15%) |
| 分类聚类 | 分类/聚类/识别/异常/聚簇 | 低(10%) |

> **国一经验**:优化和预测类获奖率最高(数据驱动 + 模型成熟)

#### 步骤 2:5 维客观打分(10 分钟)

对每道题按以下 5 维打分(每维 1-10 分):

| 维度 | 打分依据 |
|------|---------|
| **D1 数据可得性** | 题目给的数据完整吗?有没有缺失/异常/需要爬虫? |
| **D3 创新空间** | 这种题型常见解法多吗?能不能做出"评委眼前一亮"? |
| **D4 工作量** | 4 天 96h 能做完吗?(含建模+求解+写论文+调格式) |
| **D5 历史获奖率** | 这种题型往年国一占比多少? |
| **D6 卡壳风险** | 有几个环节可能卡住?(数据清洗/模型收敛/公式推导) |

总分 = D1×0.25 + D3×0.20 + D4×0.20 + D5×0.20 + D6×0.15

> **权重理由**:D1(数据坑最致命)× D3(创新是国一核心)× D4(做不完=0分)

#### 步骤 3:利弊清单 + 模型推荐(10 分钟)

对 TOP3 输出:

```markdown
### 题 X:【标题】
**总分**: 8.5/10
**题型**: 优化类
**推荐指数**: ⭐⭐⭐⭐⭐

**优势**:
- ✅ 数据完整(题目给出 3 个 CSV,xxx 行)
- ✅ 创新空间大(可结合 XXX 算法)
- ✅ 工作量适中(建模 8h + 求解 12h + 写论文 24h)

**风险**:
- ⚠️ 模型收敛可能不稳(建议用 scipy + 多起点)
- ⚠️ 优化目标可能非凸(需要松弛 + 启发式兜底)

**推荐模型**(按推荐度):
1. 【主模型】整数规划 + 拉格朗日松弛
2. 【备选 1】遗传算法 + 局部搜索
3. 【备选 2】强化学习(Q-Learning)

**关键文献**:[3 篇 arxiv 链接]
```

#### 步骤 4:决策建议

- 如果 TOP1 比 TOP2 高 ≥ 1.5 分 → 强烈推荐 TOP1
- 如果差距 < 1.0 分 → 列出"再读 30 分钟"清单,让团队定
- 如果 6 道题都 < 6 分 → 建议换题(但 4 天赛制下风险大)

### 离线兜底(无 LLM 时)

调用 `python tools/score.py --offline` 进入启发式打分模式:
- 基于关键词命中(如"最大"+"优化"→ 优化类 +1.0)
- 基于题目长度(>500 字通常数据多 +0.5)
- 基于术语密度(专业术语多 → 数据可得性高)

### 输出格式

```json
{
  "contest_id": "2026-cumcm",
  "decision_at": "2026-09-10T18:30:00",
  "rankings": [
    {"problem_id": "C", "title": "...", "score": 8.5, "category": "优化类", "recommendation": "strong"},
    {"problem_id": "A", "title": "...", "score": 7.8, "category": "预测类", "recommendation": "strong"},
    {"problem_id": "F", "title": "...", "score": 7.2, "category": "评价类", "recommendation": "consider"}
  ],
  "top3_details": {
    "C": {"pros": [...], "cons": [...], "models": [...], "risks": [...]}
  },
  "human_decision_needed": true
}
```

### 约束

- **不替团队做最终决定**——人类保留判断权
- **不打感情分**——不根据"看起来有意思"加分
- **诚实暴露不确定性**——D5 历史获奖率基于公开数据,赛后需校准
- **30 分钟硬性限制**——超过 30 分钟工具未输出 = 工具失败

Attribution

FOURTEEN1416FOURTEEN1416
View sourceSee grades on GitHubMore from FOURTEEN1416 →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →