Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

/karpathy-check — measured AI agent skill | DecimalAI

alirezarezvani/karpathy-check

NewSkillScoreMaintained

Run Karpathy's 4-principle review on staged changes or the last commit. Checks complexity, diff noise, hidden assumptions, and goal verification. Usage /karpathy-check [--last-commit]

Code Reviewby alirezarezvani

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Coding pack

decimalai skills pull alirezarezvani-karpathy-check

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/alirezarezvani-karpathy-check/SKILL.md

Raw file: /s/alirezarezvani-karpathy-check/SKILL.md ↗ · machine-readable summary at /s/alirezarezvani-karpathy-check.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

25 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 4h ago

Yes — +64% lift-40% tokens0% turns88% of whole cases pass with skill (22/25)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-07	✗→✓	▲ Improved	-14%	0%
case-08	✗→✓	▲ Improved	-65%	0%
case-09	✗→✓	▲ Improved	-66%	0%
case-10	✗→✓	▲ Improved	-58%	0%
case-20	✗→✓	▲ Improved	-57%	0%

Showing 5 of 25 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

alirezarezvani / claude-skills

.gemini/skills/karpathy-check/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1

Skill contents

SKILL.md

/karpathy-check

Review your staged changes (or last commit) against Karpathy's 4 coding principles.

Usage

/karpathy-check                 # review staged changes
/karpathy-check --last-commit   # review the most recent commit

What it runs

Principle #2 (Simplicity): engineering/karpathy-coder/skills/karpathy-coder/scripts/complexity_checker.py on all changed files — detects over-engineering, premature abstractions, deep nesting, long functions
Principle #3 (Surgical): engineering/karpathy-coder/skills/karpathy-coder/scripts/diff_surgeon.py on the diff — detects comment-only changes, whitespace noise, style drift, drive-by refactors
Principles #1 + #4 (Think + Goals): The karpathy-reviewer agent reads the diff and applies human-judgment checks — hidden assumptions, missing verification

Output

A structured report with per-principle verdicts and specific line-level fix recommendations.

When to run

Before committing (catches noise and overcomplication early)
After completing a feature (sanity check before PR)
When you suspect the LLM overcoded something

Sub-agent

Dispatches the karpathy-reviewer agent. See agents/karpathy-reviewer.md.

Scripts

engineering/karpathy-coder/skills/karpathy-coder/scripts/complexity_checker.py
engineering/karpathy-coder/skills/karpathy-coder/scripts/diff_surgeon.py
engineering/karpathy-coder/skills/karpathy-coder/scripts/assumption_linter.py
engineering/karpathy-coder/skills/karpathy-coder/scripts/goal_verifier.py

Skill Reference

→ engineering/karpathy-coder/skills/karpathy-coder/SKILL.md

Want to see if alirezarezvani/karpathy-check would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.