一键在 Manus 中运行任何 Skill

cumcm-2026-rules-watch

星标6

分支0

更新时间2026年7月16日 04:00

Track, verify, and apply 2026 CUMCM / 高教社杯 official rules, notices, paper format requirements, AI-tool rules, schedule changes, and submission instructions. Use when asked to check latest contest rules, update repository guidance for 2026 CUMCM, compare templates with official requirements, or keep preparation work aligned with current CUMCM notices.

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

bosprimigenious

bosprimigenious/ModelingPaperKit

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

CUMCM 2026 Rules Watch

Use this skill to keep ModelingPaperKit aligned with official 2026 CUMCM requirements.

Workflow

Treat official CUMCM sources as primary. Browse when the request depends on current rules, dates, notices, or AI policy.
Record exact dates and source URLs when reporting any rule.
Compare current guidance against:
- templates/cumcm/main_cumcm.tex
- templates/cumcm/sections/cover.tex
- templates/cumcm/sections/numbering_page.tex
- templates/cumcm/sections/references.tex
- README.md
- skills/modeling-paperkit/references/cumcm.md
If changing files, keep edits narrow and explain which official requirement each edit supports.
Read references/checklist.md when performing a rule audit.
Read references/official-sources.md when deciding which sources are authoritative.
Read references/examples.md when shaping a recurring rules-monitoring task.

Guardrails

Do not rely on old forum posts or copied templates as authoritative.
Do not browse public discussion of live contest problems during the active contest window unless the user explicitly asks and contest-safety rules allow it.
Do not claim the template is submission-ready unless build and submission checks have been run or the missing checks are clearly disclosed.

同仓库更多 Skills

同仓库

cumcm-baseline-models

bosprimigenious/ModelingPaperKit

Design simple, defensible baseline models for CUMCM prediction, classification, evaluation, optimization, simulation, and ranking tasks before using stronger methods. Use when a team needs a minimum viable model, method comparison, ablation baseline, or fallback path under contest time pressure.

2026-07-186

cumcm-contest-safety-mode

bosprimigenious/ModelingPaperKit

Enforce contest-time safety for CUMCM work, including no public problem discussion browsing, no outside solution sharing, source citation discipline, AI-use logging, instructor-contact boundaries, and submission-risk warnings. Use during the active CUMCM contest window or when handling current contest problem statements.

2026-07-186

cumcm-data-quality-audit

bosprimigenious/ModelingPaperKit

Audit CUMCM datasets for missing values, units, duplicates, outliers, inconsistent identifiers, impossible ranges, time/order problems, leakage, and preprocessing decisions. Use when reviewing raw or processed data before modeling, writing data cleaning notes, or deciding whether results are trustworthy.

2026-07-186

cumcm-experiment-log

bosprimigenious/ModelingPaperKit

Maintain CUMCM experiment records for datasets, preprocessing versions, model parameters, random seeds, metrics, generated figures, result tables, and paper claims. Use when tracking runs, comparing model variants, reproducing outputs, or preparing appendix and supporting-material logs.

2026-07-186

cumcm-problem-reading

bosprimigenious/ModelingPaperKit

Parse and decompose 2026 CUMCM problem statements into subquestions, data requirements, deliverables, hidden constraints, scoring signals, assumptions to verify, and first-day action plans. Use when reading a new CUMCM problem, extracting tasks, clarifying what must be modeled, or building the initial problem brief before choosing methods.

2026-07-186

cumcm-result-consistency

bosprimigenious/ModelingPaperKit

Check consistency between CUMCM paper claims, formulas, tables, figures, units, code outputs, appendix records, and subproblem conclusions. Use when reviewing a draft for mismatched numbers, stale figures, inconsistent symbols, unsupported conclusions, or broken result narrative.

2026-07-186

name	cumcm-2026-rules-watch
description	Track, verify, and apply 2026 CUMCM / 高教社杯 official rules, notices, paper format requirements, AI-tool rules, schedule changes, and submission instructions. Use when asked to check latest contest rules, update repository guidance for 2026 CUMCM, compare templates with official requirements, or keep preparation work aligned with current CUMCM notices.

CUMCM 2026 Rules Watch

Use this skill to keep ModelingPaperKit aligned with official 2026 CUMCM requirements.

Workflow

Treat official CUMCM sources as primary. Browse when the request depends on current rules, dates, notices, or AI policy.
Record exact dates and source URLs when reporting any rule.
Compare current guidance against:
- templates/cumcm/main_cumcm.tex
- templates/cumcm/sections/cover.tex
- templates/cumcm/sections/numbering_page.tex
- templates/cumcm/sections/references.tex
- README.md
- skills/modeling-paperkit/references/cumcm.md
If changing files, keep edits narrow and explain which official requirement each edit supports.
Read references/checklist.md when performing a rule audit.
Read references/official-sources.md when deciding which sources are authoritative.
Read references/examples.md when shaping a recurring rules-monitoring task.

Guardrails

Do not rely on old forum posts or copied templates as authoritative.
Do not browse public discussion of live contest problems during the active contest window unless the user explicitly asks and contest-safety rules allow it.
Do not claim the template is submission-ready unless build and submission checks have been run or the missing checks are clearly disclosed.