Math Proof
Two skills for hard mathematics problems, each ending in a self-contained proof.md: /math-proof:solo has the...
CCCXGBCUPi
SkillMaxx / Other / Math Olympiad
Math Olympiad
Solve competition math (IMO, Putnam, USAMO) with adversarial verification that catches what self-verification misses. Fresh-context verifiers attack proofs with specific failure patterns. Calibrated abstention over bluffing.
- Last updated
- Hands-on tests
- Tested ✓ in 2 of 5
Install
Tested ✓✓ Verified from docsOfficial docs ↗
/plugin install math-olympiad@claude-plugins-officialTested ✓ on 2026-10-16: installed and loaded in a throwaway sandbox with no credentials.
Log excerpt
Installing plugin "math-olympiad@claude-plugins-official"...✔ Successfully installed plugin: math-olympiad@claude-plugins-official (scope: user)
$ verify:
❯ math-olympiad@claude-plugins-official
Status: ✔ enabled1 run, latest 2026-10-16
Not yet tested
No official Codex instructions yet. We haven't found docs saying this works in Codex.
Know it works? Tell us →Tested ✓✓ Verified from docsOfficial docs ↗
Auto-loaded from your Claude Code plugins (no extra setup per xAI docs)Tested ✓ on 2026-10-16: installed and loaded in a throwaway sandbox with no credentials.
Log excerpt
Installed 1 plugin(s) from claude-plugins-official: math-olympiad $ verify: math-olympiad-c8630d86: math-olympiad [local: /home/sbx/.grok/marketplace-cache/783232b622f8182e/plugins/math-olympiad] (claude-plugins-official)
1 run, latest 2026-10-16
Not yet tested
No official Cursor instructions yet. We haven't found docs saying this works in Cursor.
Know it works? Tell us →Not yet tested
No official Pi instructions yet. We haven't found docs saying this works in Pi.
Know it works? Tell us →README
View on GitHub ↗math-olympiad
Competition math solver with adversarial verification.
The problem
Self-verification gets fooled. A verifier that sees the reasoning is biased toward agreement. arXiv:2503.21934 ("Proof or Bluff") showed 85.7% self-verified IMO success drops to <5% under human grading.
The approach
- Context-isolated verification: verifier sees only the clean proof, never the reasoning trace
- Pattern-armed adversarial checks: not "is this correct?" but "does this accidentally prove RH?" / "extract the general lemma, find a 2×2 counterexample"
- Calibrated abstention: says "no confident solution" rather than bluff
- Presentation pass: produces clean LaTeX/PDF after verification passes
Validation
17/18 IMO+Putnam 2025 problems solved, 0 false positives, 2 novel proofs found. See the skill's eval data in the anthropic monorepo.
Install
/plugin install math-olympiad@claude-plugins-official
Use
> Solve this IMO problem: [statement]
The skill auto-triggers on "IMO", "Putnam", "olympiad", "verify this proof", etc.
Preview from the source repo. Images are removed.
Similar alternatives
Nightvision
Skills for working with NightVision, a DAST and API Discovery platform that finds exploitable vulnerabilities...
CCCXGBCUPi
Mapbox
Mapbox skills and MCP servers for building location-aware applications with AI. Includes geospatial tools,...
CCCXGBCUPi
Remember
Continuous memory for Claude Code. Extracts, summarizes, and compresses conversations into tiered daily logs....
CCCXGBCUPi