Math Verifier logo

Math Verifier

Community
capwitf
math-verifier

Use when 高教社杯/CUMCM 数学建模方案需要在代码、图表、摘要或最终提交前独立核验公式、单位、量纲、约束、边界情形、可行性措辞、假设或论文就绪数学结论。

Overview

Publishercapwitf
RepositoryMy-MathModeling-skills
Skill namemath-verifier
Stars
57
Forks
1
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by capwitf on GitHub. Read the source before you install it.

Installation

Install the Math Verifier AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/capwitf/My-MathModeling-skills/tree/main/math-verifier
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/capwitf/My-MathModeling-skills.git /tmp/My-MathModeling-skills
mkdir -p .claude/skills
cp -r /tmp/My-MathModeling-skills/math-verifier .claude/skills/math-verifier
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Math Verifier in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Math Verifier on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Math Verifier is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

数学核验

用途

Verify the math before it becomes paper language. Treat this skill as an independent hard gate for dimensional consistency, formula back-substitution, boundary behavior, feasibility claims, and assumption loss.

核心规则:不要解新问题,也不要发明缺失的模型部件。只按题面、建模交接、符号表和证据产物核验已经写出、算出或声称的内容。

国一候选门槛:数学核验必须保证题目贴合、建模洞察、证据可信、可复现、边界清楚有公式、单位、约束和边界案例支撑;这是提交质量门槛,不承诺获奖。漂亮结论只要无法回代、无法闭合单位或无法说明边界,就不能进入论文就绪状态。

硬门禁:单位、定义域、约束、边界条件、公式回代或可行性措辞只要未检查、不一致或未知,模型结论就不能进入论文就绪状态。

返回规则:本技能只把发现返回 math-hub。可以建议修复面,但不能直接路由、派发或调用另一个技能。

必要输入

Collect the smallest available set:

  • current problem/subquestion and required deliverables;
  • problem_brief.md, model_handoff.md, assumption_log.csv, and symbol_table.csv when available;
  • formulas, constraints, objective, parameter values, units, and domains;
  • result tables, result_registry.csv, run_record.csv, and claim_ledger.csv for computed claims;
  • official or user-provided tolerances, hard constraints, and feasibility definitions.

If a needed source is missing, mark the affected check blocked or diagnostic-only; do not fill the gap from memory.

核验检查

Run only checks relevant to the active claim or model route:

Check typeWhat to verify
dimensionalEvery term in an equation has compatible dimensions and declared units.
domainVariables, indices, sets, binary/slack roles, and parameter ranges match the problem and code handoff.
constraintHard constraints are represented, measurable, and not silently softened.
formula back-substitutionSubstitute baseline, boundary, or reported values into formulas to catch algebra/sign/unit errors.
boundaryExtreme, zero, saturated, or limiting cases behave in the expected direction.
conservationFlow, mass, count, probability, budget, or balance laws close within tolerance.
feasibility wordingClaims use strict feasible, relaxed diagnostic, candidate, not identifiable, or unavoidable gap correctly.
assumption lossSimplifications name what realism, constraint, or uncertainty they remove.
result schemaThe table/figure needed to prove the claim can hold all required units, scenarios, baselines, and tolerances.

不要奖励复杂度。单位闭合、边界已检查的简单模型,优先于隐藏可行性缺口的复杂模型。

产物约定

When maintaining math_verification.csv, use:

csv
check_id,subquestion,artifact,location,claim_id,check_type,input_ref,expected_relation,observed,status,severity,minimum_fix,owner

Statuses:

  • pass: check is directly satisfied by cited evidence.
  • blocked: required source, unit, formula, constraint, or tolerance is missing.
  • fail: evidence contradicts the model or claim.
  • diagnostic-only: check reduces uncertainty but is not enough for paper-ready use.

Severity:

  • P0: contradiction, fabricated value, missing required deliverable, or rule-breaking feasibility wording.
  • P1: unit, constraint, boundary, reproducibility, or validation gap blocking paper-ready status.
  • P2: weak assumption explanation, missing stress case, or fragile boundary claim.
  • P3: wording, notation, or presentation cleanup after the math is sound.

输出约定

Return:

text
Verification lock:
Checked artifacts:
Hard check results:
P0/P1 blockers:
Claims allowed as paper-ready:
Claims downgraded to candidate/diagnostic:
Assumption and boundary losses:
Math verification rows:
Smallest repair order:
Return to hub: math-hub
Recommended repair surface:

Recommended repair surface may name model, code, table, figure, abstract, latex, or hub QC, but the handoff always returns to math-hub.

红线

Stop and mark a blocker when:

  • a symbol has no unit, domain, or first-use definition;
  • two terms are added or compared with incompatible dimensions;
  • a hard constraint is absent from the model, code, or feasibility check;
  • a reported result cannot be reproduced by formula back-substitution or a cited run;
  • a relaxed, violated, or candidate solution is described as strict feasible;
  • a boundary case contradicts the claimed mechanism or trend;
  • an assumption removes a real constraint without naming the loss;
  • verification starts designing a new route instead of checking the current one.

Do not hide failed checks in prose. Downgrade the claim or return the missing evidence to math-hub.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Math Verifier AI skill do?

Use when 高教社杯/CUMCM 数学建模方案需要在代码、图表、摘要或最终提交前独立核验公式、单位、量纲、约束、边界情形、可行性措辞、假设或论文就绪数学结论。

Why use Math Verifier on TypingMind?

Because you install it once and use it with any model. Math Verifier is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Math Verifier in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/capwitf/My-MathModeling-skills/tree/main/math-verifier. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Math Verifier?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Math Verifier?

As many as you like. As long as a model supports skills, you can use Math Verifier with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Math Verifier AI skill free?

Yes. It is published on GitHub by capwitf under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇