Back to directory

MathModel Skill Standard

A set of workflow skills for coding agents such as Trae, Claude Code and Codex that walks a mathematical-modeling competition entry from problem analysis through real computation to a finished, checked paper.

MathModel Skill Standard

MathModel Skill Standard gives a coding agent a repeatable workflow for mathematical-modeling competitions: analyze the problem statement and attachments, choose a model, write and run code, organize the resulting charts and evidence, then write and check the paper, rather than asking the model to generate an essay directly.

The project ships four tiers on separate branches for different time and compute budgets: Flash for a fast basic draft, Lite for a low-effort basic report, Standard for a complete competition paper with citation and native Word or PDF checks, and Pro for the highest-effort version with multiple comparison routes and a five-role review. An experimental LaTeX branch offers a TeX and PDF export path instead. Each competition project installs exactly one version's package into its agent's skill directory, and the release archive is verified against a published checksum.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.