Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add hamzafarooq/claude-code-starter --skill karpathy-reviewgit clone --depth 1 https://github.com/hamzafarooq/claude-code-starterWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/hamzafarooq/claude-code-starter/karpathy-review)<a href="https://agentmods.dev/skills/hamzafarooq/claude-code-starter/karpathy-review"><img src="https://agentmods.dev/badge/skills/hamzafarooq/claude-code-starter/karpathy-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/hamzafarooq/claude-code-starter/karpathy-review"><img src="https://agentmods.dev/badge/skills/hamzafarooq/claude-code-starter/karpathy-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00058 | $0.00517 |
| Opus 5.5 | $0.00023 | $0.00207 |
| Sonnet 5.5 | $0.00012 | $0.00103 |
| Haiku 4.5 | $0.00006 | $0.00052 |
Grade A, and why
karpathy-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Review the provided code, plan, or spec against Karpathy's four principles. Credit: guidelines distilled from Andrej Karpathy's January 2026 observations, adapted by Forrest Chang.
Ask the user: "Paste the code, plan, or spec you want reviewed."
Then score it against each principle:
Karpathy Review
1. Think Before Coding — were assumptions stated?
- Did the solution state its assumptions explicitly, or pick silently between interpretations?
- Are there any unclear requirements that should have triggered a clarifying question?
- Finding: PASS / FLAG — [specific observation]
2. Simplicity First — is this the minimal solution?
- Does it add features beyond what was asked?
- Are there abstractions written for single-use code?
- Is there "flexibility" or "configurability" that wasn't requested?
- Could this be meaningfully shorter?
- Finding: PASS / FLAG — [specific observation, e.g. "This 180-line class could be a 20-line function"]
3. Surgical Changes — does it stay in its lane?
- Does it modify adjacent code, formatting, or comments that weren't part of the task?
- Does it refactor things that weren't broken?
- Does it delete pre-existing code that wasn't asked to be removed?
- Finding: PASS / FLAG — [specific observation]
4. Goal-Driven Execution — is success verifiable?
- Is there a clear success criterion?
- For multi-step work: was a plan stated with checkpoints?
- Is there a way to verify the output without running it?
- Finding: PASS / FLAG — [specific observation]
Verdict
| Principle | Result |
|---|---|
| Think before coding | ✅ PASS / ⚠️ FLAG |
| Simplicity first | ✅ PASS / ⚠️ FLAG |
| Surgical changes | ✅ PASS / ⚠️ FLAG |
| Goal-driven execution | ✅ PASS / ⚠️ FLAG |
One thing to fix: [If any flags, name the single most important change to make]
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 51 lines · 58 tokens per session scan A d14141f5dc4b
karpathy-review is a skill published in the GitHub repository hamzafarooq/claude-code-starter (145 stars, last pushed 7d ago), licensed MIT. It adds 58 tokens to every session and 517 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…