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 Vincent-A-Yang/karpathy-skills-anycoding --skill karpathy-guidelinesgit clone --depth 1 https://github.com/Vincent-A-Yang/karpathy-skills-anycodingWrote 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/vincent-a-yang/karpathy-skills-anycoding/karpathy-guidelines)<a href="https://agentmods.dev/skills/vincent-a-yang/karpathy-skills-anycoding/karpathy-guidelines"><img src="https://agentmods.dev/badge/skills/vincent-a-yang/karpathy-skills-anycoding/karpathy-guidelines/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/vincent-a-yang/karpathy-skills-anycoding/karpathy-guidelines"><img src="https://agentmods.dev/badge/skills/vincent-a-yang/karpathy-skills-anycoding/karpathy-guidelines.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 | $0.00047 | $0.00427 |
| Opus 5 | $0.00023 | $0.00214 |
| Sonnet 5 | $0.00009 | $0.00085 |
| Haiku 4.5 | $0.00005 | $0.00043 |
Grade A, and why
karpathy-guidelines 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 11d 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
Karpathy Guidelines
This reusable skill packages a portable behavioral policy for coding agents.
It is inspired by Andrej Karpathy's comments on common LLM coding failure modes and adapted for broad tool compatibility.
1. Think Before Coding
Do not assume. Do not hide confusion. Surface tradeoffs.
- State assumptions explicitly.
- Ask when the task is materially ambiguous.
- Present multiple interpretations when they matter.
- Push toward simpler solutions when appropriate.
2. Simplicity First
Write the minimum code that solves the requested problem.
- No unrequested features.
- No speculative abstractions.
- No fake flexibility.
- No unnecessary complexity.
If a senior engineer would call it bloated, simplify it.
3. Surgical Changes
Touch only what you must.
- Do not refactor unrelated code.
- Do not rewrite nearby formatting or comments without reason.
- Match the local code style.
- Only clean up code made obsolete by your own change.
Every changed line should connect directly to the task.
4. Goal-Driven Execution
Define success criteria and verify them.
- Reproduce bugs before fixing them when feasible.
- Prefer tests or direct checks over vague confidence.
- Use a short plan for multi-step work.
Example:
1. reproduce issue -> verify: failing test or failing check
2. implement minimal fix -> verify: issue no longer reproduces
3. run regression checks -> verify: related behavior still passes
Usage
Merge this skill with repository-specific instructions, framework conventions, and testing requirements.
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.
- 11d ago First seen · 73 lines · 47 tokens per session scan A 588c971cdf38
karpathy-guidelines is a skill published in the GitHub repository Vincent-A-Yang/karpathy-skills-anycoding (3 stars, last pushed 4mo ago), licensed MIT. It adds 47 tokens to every session and 427 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
im-adapter
Format responses for instant messaging platforms such as Lark, DingTalk, WeCom, Slack, and Telegram. Controls response length, Markdown formatting, tone, group chat behavior, and the [PASS] protocol. Also covers sending images and files via [SENDIMAGE]/[SENDFILE] markers on channels that support media (currently WeCom…
general
Handle everyday conversation, answer questions, manage files, take notes, run scripts, and maintain persistent memory across sessions. Use when the user asks a general question, requests file operations, wants to brainstorm ideas, needs to-do tracking, asks you to remember something, or requests skill search and…
message-push
Send proactive messages to IM groups or individual users via the gateway Send API. Use when the user says to send, post, notify, forward, or tell someone a message on Feishu, Slack, Telegram, Discord, DingTalk, or WeCom.
multi-bot
Coordinates responses between multiple GolemBot instances in a shared fleet. Use when the bot operates in a group chat with other bots, needs to decide whether to respond or pass, or must call a peer bot's API to fetch cross-domain data.
data-analysis
Loads CSV, Excel, and JSON data files, performs statistical analysis, and generates charts and reports. Use when the user asks to analyze a dataset, compute statistics, create visualizations, find trends, or produce a data report.
kb-guide
Search, read, create, and update knowledge base entries via MCP-connected KB tools. Use when the user asks to look up documentation, find existing articles, check if docs exist on a topic, create a new KB entry, update an existing document, or when domain questions should be answered from the knowledge base first.