read-problem
Skill9 starsChanged 4 months agoMITView on GitHub
解析当年比赛题目(读题/审题)。当用户把官方题目 PDF/文本放进 inputs/problem/ 并要"读题""审题""解析题目""提取题目要求/限制条款"时使用。产出结构化 design/problem.yaml(限制条款、要求功能、评分、场地、计时),作为后续方案/选材的唯一需求来源。
In this repository11 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- boardSkill
- elec-designSkill
- firmware-scaffoldSkill
- interconnectSkill
- plan-solutionSkill
- power-designSkill
- select-partsSkill
- setup-envSkill
- test-checklistSkill
- vision-scaffoldSkill
The file
Files
read-problem —— 读题/审题(流水线入口 ①)
lane: lead · needs: multimodal(解析 PDF/图片 → 只在 lead/Claude 会话跑;队友 lane 直接用产出的 problem.yaml)
这是整条流水线的真正入口。每年题目规定都不同,一切从读懂当年题目开始——尤其先看「限制条款」(如允不允许用摄像头,直接决定 K230 上不上)。
步骤
- 定位题目:在
inputs/problem/找当年题目文件(PDF/文本/图片)。没有就提醒用户投喂,停。 - 读全文:用 Read 读题目(PDF 多模态直接读)。逐条抠:基础要求、发挥部分、评分、场地尺寸/标记、时间限制、器件/重量/尺寸限制。
- 查 KB 作背景(用 Read,不复制):
kb/01-赛事概况与赛制流程.md—— 赛制、限制条款的常见形态、评分惯例。kb/02-历年小车与控制类赛题.md—— 对照历年同类题型,识别"这是哪一类"(送药/追踪/循迹/双车/识别)。
- 抽成结构,写
design/problem.yaml(schema 见下)。限制条款单列、最显眼。 - 标注模糊点:题目没写清、或需要现场确认的(场地材质、起停判定、是否允许外部供电等),放进
ambiguities,提请用户确认——这是轻人工门,别替用户拍板硬约束。 - 更新
STATUS.md:阶段 → ②;lead 段标"题目已解析,待 /plan-solution"。
design/problem.yaml schema
title: # 题目名
year: # 年份/届
category: # 送药 | 追踪 | 循迹 | 双车 | 识别 | 其他(对照 KB02)
source: # inputs/problem/<文件名>
restrictions: # ★ 限制条款——最先看、最关键
camera_allowed: # true | false | unknown(决定 K230)
power: # 供电限制(如禁外接电源/电压上限)
size_weight: # 尺寸/重量限制
parts: # 指定/禁用器件
other: []
required_functions: # 要实现的功能
- {id: F1, desc: , level: basic|advanced, points: }
scoring: # 评分细则
- {item: , points: }
arena: {desc: , dimensions: , track: , markings: }
timing: {run_limit: , notes: }
constraints: [] # 其他硬约束
ambiguities: # 需人工确认(不要自己默认)
- {q: , why: }注意
- 不要发明题目没写的东西;题目没明确的进
ambiguities,别填默认值(默认值正是赛场翻车点)。 - 限制条款里
camera_allowed=false时,要在产出里显著提示:本届不能上 K230,方案需走纯灰度/电磁,后续plan-solution据此选。 - 产出是给
plan-solution/select-parts用的机器可读需求,写清楚、可签字。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

