MathModel Copilot skill
A skill for mathematical modeling contests: from problem analysis to verified results, with a workbench for a three-person team.
- 67 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The skill runs Python calculation code and writes state and report files in the project workspace.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/mathmodel-copilotDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skill runs Python calculation code and writes state and report files in the project workspace.
This is third-party code. Review the repository files before installing.
What it does
MathModel Copilot guides a team through a mathematical modeling contest (CUMCM, MCM/ICM, Diangong and custom formats). The skill breaks down requirements, models each sub-question, runs real calculations in Python and verifies results independently. Project state lives in a versioned structure, and the `copilot.py` command creates a workspace and shows status and context for a modeler, coder or writer role. A local modeling workbench shows project progress and AI feedback, while shared materials give all three team members and the agent one picture. It also helps with the paper text, checks the submission package before delivery and has a teaching mode that explains the skill step by step. The base is mathmodel-skill version 6.2.0 with stages 0–9.
Who it is for. Student modeling teams and coaches who want to run a contest together with an agent while keeping decisions in human hands.
Good fit when
- A team received the problem and attachments and wants to analyze requirements and choose an approach.
- Calculations must run in Python and be verified separately from the model.
- Three members share a project and need one status view and context handoff across sessions.
- Before submission the package must be checked and version mixing avoided.
Not a fit when
- You want a finished paper generated automatically with no human involvement; the skill is built around joint work.
- Your contest rules forbid AI use during the competition.
- The task is outside mathematical modeling.
Example request
Install MathModel Copilot, create a CUMCM project and show the status before starting.Limitations
The skill is marked version 0.3 in Preview, so behavior and the state schema may still change. Documentation and the main interface language are Chinese. Word, MATLAB, network services and parallel-agent capabilities are detected on site and unconfirmed ones count as unknown. Check that your contest rules allow AI help before using it in a real competition. The skill reads and writes project files, so run it in a dedicated workspace. Licenses of different parts are described in `LICENSE_SCOPE.md`.
How to disable. Remove the skill folder from your agent's skills directory. The project workspace with state and reports stays on disk and is deleted separately.
FAQ
How do I install the skill?
Download the archive from the repository, send it to your agent and ask it to install the skill. Alternatively, unpack it into the skills directory by hand.
Which contests are supported?
CUMCM, MCM/ICM, Diangong and custom formats through the `generic` option of `init`.
What does teaching mode do?
It explains the skill's features step by step and guides a user with no task yet. Start it by asking to enter teaching mode.
How do I resume in a new session?
Run `status` and `context --role <role> --member <name>`; they show shared facts, accumulated changes, tasks and blockers.
How does the skill verify results?
Calculations run in Python, checks are independent of the model and completion status ties to versioned requirements and actual evidence.
Discussion
No questions or reviews yet.
Related
SkillsA skill for mathematical modeling contests: from problem analysis and model comparison to a verified LaTeX paper and PDF.
Autoresearch for skills
autoresearch
A skill that improves another skill: runs it on test inputs, scores outputs with binary checks and mutates the prompt
A skill with 25 commands to audit and improve the prompts, tools and agent designs of a project in Cursor, Claude Code and Codex
A skill that turns several Claude Code, Codex and Cursor windows into a team with a coordinator, shared registry and cross-review
Ollaya skill and MCP server: an agent puts typed questions to local decision models and gets probabilities with no LLM call
A desktop app where Claude Code, Codex, Gemini and other agents work as a team of specialists in a shared project
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| MathModel Copilot (this card) | no | 67 | |
| mathmodel-skill | no | 298 | the base ten-stage skill for a student team; Copilot adds versioned state, a workbench and team roles |
| Math Modeling Skills | no | 437 | two skills for CUMCM and MCM/ICM covering problem breakdown and paper; this one is a shared workflow with result verification |
README in short
The skill provides a stage 0–9 workflow with per-sub-question loops, a competition pack, a critic and red team, initialization and paper assembly. The repository includes a Codex plugin, templates and license scope notes.