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 IrtezaAsadRizvi/article-writing-skills --skill andrej-karpathygit clone --depth 1 https://github.com/IrtezaAsadRizvi/article-writing-skillsWrote 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/irtezaasadrizvi/article-writing-skills/andrej-karpathy)<a href="https://agentmods.dev/skills/irtezaasadrizvi/article-writing-skills/andrej-karpathy"><img src="https://agentmods.dev/badge/skills/irtezaasadrizvi/article-writing-skills/andrej-karpathy/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/irtezaasadrizvi/article-writing-skills/andrej-karpathy"><img src="https://agentmods.dev/badge/skills/irtezaasadrizvi/article-writing-skills/andrej-karpathy.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 | $0.00092 | $0.02816 |
| Opus 5 | $0.00046 | $0.01408 |
| Sonnet 5 | $0.00018 | $0.00563 |
| Haiku 4.5 | $0.00009 | $0.00282 |
Grade A, and why
karpathy-article-writing 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 12d 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.
How it starts
The opening of the file, as written. The whole thing — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Writing in Karpathy's Style
This skill distills the voice found across four reference posts:
- A from-scratch tour of the Bitcoin protocol in Python (2021) — a dense technical tutorial
- Short story on AI ("Forward Pass", 2021) — speculative fiction from a model's POV
- Biohacking Lite (2020) — a year-long self-experiment written up
- Deep Neural Nets: 33 years ago and 33 years from now (2022) — a paper reproduction + retrospective
The voice is the same across all four despite the wildly different genres. That voice is what this skill captures.
The five-word north star
Curious. Built it. Write it up.
Karpathy doesn't opine; he makes things and then reports back on what he found. If the piece isn't grounded in something the author actually did — implemented, reproduced, measured, lived through — the voice falls apart. Before writing, make sure there's a concrete artifact (code, data, a lived experiment, a reproduction) at the center.
Voice rules
First person, always
- "I" for the author's actions, opinions, hedges: I set out to reproduce…, I suspect…, I was a bit sketched out about…
- "We" for collaborative walkthroughs where the reader is coming along: we are going to…, we now have…
- Avoid passive constructions. The weights were initialized → I initialized the weights (or we initialize the weights).
Conversational, but never dumbed down
- Use domain jargon without defensive scaffolding. UTXO, ECDSA, MACs, glycogen, oxidative phosphorylation — drop them in, gloss only if they're genuinely load-bearing for the next sentence.
- The reader is assumed smart and curious. Trust them.
- No "let me explain X in simple terms" framing. Just explain it.
Intellectual honesty via hedges
Karpathy hedges constantly, and it reads as honesty rather than weakness because the hedges are about things genuinely uncertain:
- I suspect…
- most likely…
- I believe the…
- probably…
- as far as I can tell…
- ¯\(ツ)/¯
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.
- 12d ago First seen · 203 lines · 92 tokens per session scan A 5f8a0ddd9c7e
karpathy-article-writing is a skill published in the GitHub repository IrtezaAsadRizvi/article-writing-skills (17 stars, last pushed 4mo ago), licensed MIT. It adds 92 tokens to every session and 2,816 once invoked, about $0.0005 per session on Opus 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 2026-08-30.
Other skills, from other repositories
dcf-model
Build discounted cash flow valuation workbooks in Excel.
grounded-citations
Ground answers and documents in cited, verifiable sources.
hermes-agent-skill-authoring
Author in-repo SKILL.md files: frontmatter and structure.
baoyu-article-illustrator
Article illustrations: type × style × palette consistency.
flash-attention
Speed up long-sequence transformer training and inference.
subagent-driven-development
Execute plans via delegatetask subagents (2-stage review).