메뉴
数学建模竞赛 V2 题面与数据建档阶段。用于读取赛题、附件和资源，拆解子问题，审计数据风险，建立可恢复上下文，并为后续建模与代码阶段提供结构化输入。
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | mm-problem-intake |
| description | 数学建模竞赛 V2 题面与数据建档阶段。用于读取赛题、附件和资源，拆解子问题，审计数据风险，建立可恢复上下文，并为后续建模与代码阶段提供结构化输入。 |
| allowed-tools | Bash(*), Read, Write, Edit, Grep, Glob, Agent, WebSearch, WebFetch |
文件关系全貌请见 [[FILE_RELATIONSHIP_MAP]] · 上游: [[skills/mm-start-contest-v2/SKILL|V2 总控]] · 下游: [[skills/mm-model-strategy/SKILL|Phase 2 Strategy]] · 共享规范: [[skills/_references/SKILL|_references]]
Read:
../_references/v2_pipeline_contract.md../_references/codex_subagent_protocol.md../_references/rag_usage_contract.md../_references/source_quality_policy.md../_references/problem_type_router.mdCreate or update:
PROBLEM_BRIEF.mdDATA_AUDIT.mdWORKFLOW_STATE.mdreports/AGENT_RUNS.mdreports/INTAKE_GATE.mdques1, ques2, etc. Do not inflate small details into fake subproblems.cumcm_problems or mcm_icm_problems for similar task wording, attachment patterns, and hidden scoring cues. Record only sourced hits using ../_references/rag_usage_contract.md and ../_references/source_quality_policy.md; core problem facts should come from the current files or S official sources, not unsourced memory.WORKFLOW_STATE.md; mark any unreadable or semantically unclear data as HIGH risk.problem-analyst and data-auditor independently, then summarize their outputs into reports/AGENT_RUNS.md.# Problem Brief
## Contest And Task
## Source Files
## Top-Level Questions
## Inputs And Outputs
## Constraints And Evaluation
## Dependencies Between Questions
## Ambiguities And Risks
## Next Modeling Needs
## Local RAG Evidence Used
# Data Audit
## File Inventory
## Table And Field Summary
## Missing / Abnormal / Duplicate Data
## Unit And Encoding Issues
## Usable Variables By Question
## Derived Variables Needed
## Blocking Risks
Write reports/INTAKE_GATE.md with PASS, CONDITIONAL_PASS, or FAIL.
Do not proceed to modeling on FAIL. On CONDITIONAL_PASS, list exactly what the model strategy stage must handle.
环境检查与安装向导。检查 MathModelAgent V2.6 工作流所需依赖是否已安装，对缺失项提供安装命令，并在用户确认后执行安装。手动触发。
数学建模竞赛 V2.6 高分论文对标评审阶段。用于按高分论文标准审查模型、实验、图表、论文结构、结论追踪和可提交性，并给出修改清单。
Mathematical modeling contest V2.6 revision loop. Use after mm-contest-review when PAPER_SCORECARD or REVISION_ACTIONS contains BLOCKER, HIGH, MEDIUM, weak-claim, figure-audit, or method-implementation issues that must be repaired before final verification.
数学建模竞赛 V2.6 最终验收阶段。用于检查完整产物、图表插入、结果追踪、论文编译、代码可复现、高分论文评分和提交就绪状态。
数学建模竞赛 V2.6 总控入口。用于启动高分论文对标的 Skill + Codex 子代理混合工作流，创建持久化上下文、调度题面拆解、建模评审、代码实验、图表、论文写作和最终验收。
共享规范知识库。包含 V2.6 流程契约、Codex 子代理协议、评分标准、图表标准、本地 RAG 使用契约、题型路由、模型卡和评审协议等参考内容。其他 skills 在执行过程中按需读取；不要单独触发本 skill。