agentic-karpathy

agentic-karpathy is a skill for Claude Code, Codex from tmdgusya/roach-pi. It costs 57 tokens per session (1,778 once invoked), scanned A, original, no licence file.

A set of coding guidelines based on computer scientist Andrej Karpathy’s practices. It encourages small changes, checked assumptions, and careful control of the task’s scope.

In plain words
What is it for?
Use it when implementing features, modifying code, or checking that a planned change is properly understood and narrowly targeted.
Why use it?
It reduces common mistakes such as editing before understanding the existing code or making unnecessary changes.

Skill for Claude CodeCodex

Which agent this was written for is unclear — body not stored (licence); the path alone says nothing.

Good fit Use it when implementing features, modifying code, or checking that a planned…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/tmdgusya/roach-pi/agentic-karpathy
Install

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.

Any agent
npx skills add tmdgusya/roach-pi --skill agentic-karpathy
Clone the repo
git clone --depth 1 https://github.com/tmdgusya/roach-pi

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for agentic-karpathy

README.md
[![agentmods](https://agentmods.dev/badge/skills/tmdgusya/roach-pi/agentic-karpathy.svg)](https://agentmods.dev/skills/tmdgusya/roach-pi/agentic-karpathy)
Your own site
<a href="https://agentmods.dev/skills/tmdgusya/roach-pi/agentic-karpathy"><img src="https://agentmods.dev/badge/skills/tmdgusya/roach-pi/agentic-karpathy.svg" alt="Measured on agentmods" height="20"></a>
Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,778 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin unknown No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00057 $0.01778
Opus 5 $0.00028 $0.00889
Sonnet 5 $0.00011 $0.00356
Haiku 4.5 $0.00006 $0.00178

Measured 7d ago against content hash 4d02958ea646, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

agentic-karpathy 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 7d 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.

extensions/agentic-harness/skills/agentic-karpathy/SKILL.md · 192 lines

The source is not reproduced here

No licence file

A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.

Read it on GitHub

Changes

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.

  1. 7d ago First seen · 192 lines · 57 tokens per session scan A 4d02958ea646

Subscribe to this mod's changes

agentic-karpathy is a skill published in the GitHub repository tmdgusya/roach-pi (275 stars, last pushed 1mo ago), with no licence file. It adds 57 tokens to every session and 1,778 once invoked, about $0.0003 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.

Related

Other skills, from other repositories

explain-pr

Walk a user through a GitHub pull request to help them understand what it does, using markdown visualisations and a guided diff-by-diff tour. Use when the user wants to understand, explore, learn from, or be walked through a PR (not when they want a formal review).

kimgoetzke/coding-agent-configs · 62 tokens

review-pr

Review a GitHub pull request for correctness, security, observability, test coverage, and conventions. Identify applicable skills, verify and triage sub-agent findings to cut noise, score each review dimension purple-red-amber-green, and produce a structured review saved to disk and presented to the user. Use when…

kimgoetzke/coding-agent-configs · 85 tokens

codebase-hygiene

Find semantic duplicates (same intent, different implementation) and shallow modules (thin wrappers spreading complexity). Read-only advisory — changes route through BUILD with full gates. Use when auditing code quality, before refactoring, or when a codebase feels bloated.

romiluz13/auto-pi · 56 tokens

security-review

Security review of the diff — the third review axis alongside Standards and Spec. Reviews for injection, auth, secrets, deserialization, SSRF, path traversal, unsafe operations, and dependency confusion. Use as a parallel sub-agent with fresh context (diff-only, anti-anchored) during the review phase.

romiluz13/auto-pi · 65 tokens

code-review

Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes: Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/spec asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use…

wangmiaozero/pi-harness · 94 tokens

improve-codebase-architecture

Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

wangmiaozero/pi-harness · 32 tokens