maths-olympiad

Solve competitive mathematics problems with multi-agent verification and LaTeX proofs.

4|Updated May 20, 2026
One-click install
npx skills add https://github.com/valorisa/Claude-Skills --skill maths-olympiad
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: maths-olympiad
Source: https://github.com/valorisa/Claude-Skills/tree/main/skills/maths-olympiad
Command: npx skills add https://github.com/valorisa/Claude-Skills --skill maths-olympiad

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and scripts (resource) components.

What problem does it solve?

Solving competitive mathematics problems (such as IMO, Putnam, or AIME) carries high risk of misinterpreting problem statements and missing subtle logical gaps in proofs, leading to incorrect solutions presented with false confidence. This Skill eliminates that risk with a structured, multi-agent workflow that prioritizes accuracy and rigorous verification over guesswork.

Core Features & Use Cases

  • Parallel Multi-Agent Solving: Launches 8-12 specialized solver agents with varied attack angles (invariants, induction, extremal cases) to explore problem spaces exhaustively.
  • Adversarial Context-Isolated Verification: Uses fresh, blind verifier agents to attack cleaned proofs for gaps, misinterpretations, and logical errors, with asymmetric voting to balance false positives and false negatives.
  • Calibrated Abstention: Provides honest "no confident solution" responses with partial progress notes instead of wrong, overconfident answers, optimizing conditional accuracy.
  • LaTeX Proof Presentation: Formats verified, correct proofs into clean, elegant LaTeX for professional or academic use.
  • Use Case: A math coach preparing students for the IMO can use this Skill to verify student solutions, catch subtle logical gaps, and generate polished, competition-ready proof writeups.

Quick Start

Use the maths-olympiad skill to solve and verify a rigorous proof for the attached Putnam problem statement, outputting the final verified proof in clean LaTeX format.

Frequently Asked Questions about maths-olympiad

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify olympiad math proofs for logical gaps?

You can verify olympiad math proofs using a workflow with adversarial verification. Fresh, blind verifier agents attack cleaned proofs to identify logical gaps and misinterpretations, utilizing asymmetric voting to balance false positives and false negatives.

What is the best way to solve competitive mathematics problems like IMO and Putnam?

The best way to solve competitive mathematics problems is parallel multi-agent solving. This approach launches 8-12 specialized solver agents with varied attack angles like induction and extremal cases to exhaustively explore the problem space.

Can I get LaTeX formatted output for verified math olympiad solutions?

Yes, you can get LaTeX formatted output for verified math olympiad solutions. The workflow formats verified, correct proofs into clean, elegant LaTeX for professional or academic use, ensuring mathematically sound presentation.

How does adversarial verification work for competitive math proofs?

Adversarial verification for competitive math proofs works through context-isolated verification. Fresh, blind verifier agents independently attack the cleaned proofs for logical errors and misinterpretations, using asymmetric voting to validate proof correctness.

What happens when a math olympiad solver cannot confidently find a solution?

When a math olympiad solver cannot confidently find a solution, it uses calibrated abstention. It provides honest no confident solution responses with partial progress notes instead of wrong, overconfident answers, optimizing for conditional accuracy.

Does the competitive math solver need web access to preserve reasoning integrity?

No, the competitive math solver requires no web access to preserve reasoning integrity. It operates completely offline to ensure the mathematical proofs are solved through context-isolated verification without relying on external data.