Lucasuiii/modeling-workbench / .agents/skills/cumcm-workflow
cumcm-workflow
VerifiedBuild or resume mathematical-modeling competition work, including CUMCM, MCM/ICM, graduate and regional contests, data challenges, and open-topic statistical modeling. Guide problem framing, modeling, computation, validation, Chinese or English LaTeX writing, and reviewed PDF/source delivery using current official requirements. Not for ordinary paper polishing.
references/02-problem-analysis.md
View on GitHub# Stage 2: Problem analysis
For a team-selected question under an official theme, first use [open-topic research](open-topic-research.md), then express the chosen research questions through the same outputs below. When observations or data preparation determine the answer, consult the data section of [task-driven modeling](task-driven-modeling.md). Keep official requirements, external observations and team assumptions distinguishable.
## Required outputs
- `analysis/PROBLEM_FACTS.json`
- `analysis/TASK_CAPABILITIES.json`
- `analysis/ASSUMPTIONS.md`
- `analysis/SYMBOLS.md`
For each fact, cite an official source file and page, sheet, table, or cell range. Separate stated facts, interpretations, and added assumptions.
Translate each subproblem into observable capabilities: requested output, official facts used, acceptance checks, intended model owners, expected code entry points, and result IDs. Capability coverage is complete only when every requested subproblem output has an owner and a check that could fail.
Write each acceptance check so a program can fail it. The failure this guards is the task quietly shrinking until it fits what the agent can already do: every check after this point verifies faithfulness to the reading that was written down, so none of them can see that the reading itself narrowed.
- `judge: "recorded"` means the solving program decides. Name the assertion in `assertion_name`; the program writes it out through `record_run.py --assert-file`, and `CAP-E012` checks that an official run for this capability actually recorded it passing. Phrase `assertion` as the condition and its negation: "the enumerated count equals the class cardinality; fewer means a policy was skipped". Reserve this for anything a program could settle — a count, a bound, a required field being present and non-empty, an invariant holding on the output.
- `judge: "human"` is for the questions no script settles: whether a reading of the problem is right, whether an interpretation is faithful. These go to the reviewer, and claim nothing.
A capability with only `human` checks is allowed and sometimes correct, but it buys no machine evidence, so prefer a recorded one wherever the delivery has any observable consequence at all.
Use stable fact and capability IDs. Do not treat a method name or keyword as implementation evidence. Identify ambiguous wording and compare plausible interpretations before choosing one.
Record the chosen interpretation and its alternatives. In `working` mode, continue with reversible model exploration when ambiguity does not change the data meaning or requested output. Ask for an explicit user decision only when competing interpretations would materially change the mathematical task, official-data use, or final claims; do not silently freeze a final model before that decision.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| canvas-designanthropics/skillsCreate beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| docxanthropics/skillsUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', '.dotx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx or .dotx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pdfanthropics/skillsUse this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pptxanthropics/skillsUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates (.potx), layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx or .potx filename, regardless of what they plan to do with the content afterward. If a .pptx or .potx file needs to be opened, created, or touched, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |