karpathy-guidelines SKILL.md skillA

karpathy-guidelines from rudycity/superagent is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add rudycity/superagent/karpathy-guidelines@git:20260702.4f76578

Pins git:20260702.4f76578 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
rudycity/superagent
Path
.agents/skills/karpathy-guidelines/SKILL.md
Versions recorded
4
Latest version
git:20260702.4f76578, committed Jul 2, 2026
First version recorded
Jun 11, 2026
Size
2,206 bytes
Repository stars
52
License
MIT

Behavioral rules to minimize common LLM coding pitfalls. Applies to all code creation, modification, review, and validation tasks.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# ROLE & PRINCIPLE
- **Goal**: Minimize LLM coding mistakes, bias toward caution and surgical precision.
- **Rule**: Never assume or guess. Keep changes isolated.

# EXECUTION ROUTING
if task_type == "plan_design" OR task_type == "new_feature":
    CALL apply_thinking_rules()
    CALL apply_simplicity_rules()
elif task_type == "bug_fix" OR task_type == "edit_existing":
    CALL apply_thinking_rules()
    CALL apply_surgical_rules()
elif task_type == "validation" OR task_type == "testing":
    CALL apply_verification_rules()

# 1. THINKING RULES
Before editing or writing code:
- State assumptions explicitly. if unsure: CALL ask_question()
- if multiple_implementations: Present trade-offs to user. Do NOT make silent choices.
- if task_unclear: Identify confusing point. CALL ask_question(). Do NOT guess.
- if simpler_approach_exists: Suggest it immediately. Push back on over-engineering.

# 2. SIMPLICITY RULES
- Write minimum code to solve requirements.
- No speculative abstractions or single-caller helper layers.
- No unrequested configurability or parameter flexibility.
- No error-handling paths for impossible/unreachable scenarios.
- if line_count > necessary: Rewrite to condense.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add rudycity/superagent/karpathy-guidelines@sha256:c661430ff159c3d5

The line at the top of this page pins the label git:20260702.4f76578, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_7eepfg33ptfw7gy4.svg)](https://markdownregistry.com/a/art_7eepfg33ptfw7gy4)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 4 versions of this file, between Jun 11, 2026 and Jul 2, 2026, about one every 7 days.

versioncommittedcommitsizeaudit
git:20260702.4f76578 latest2026-07-02 4f76578 2,206 BA view · diff
git:20260623.12275dd2026-06-23 12275dd 2,517 BA view · diff
git:20260623.c6692d92026-06-23 c6692d9 2,323 BA view · diff
git:20260611.23f00652026-06-11 23f0065 2,304 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (2206 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

rudycity/superagent · 52 stars · license MIT · pushed 2026-10-20 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_7eepfg33ptfw7gy4
GET https://markdownregistry.com/api/v1/resolve?ref=rudycity/superagent/karpathy-guidelines
GET https://markdownregistry.com/api/v1/blob/c661430ff159c3d53ed8f36d8ab992d1af058692d8658d8616c0ffd170e16fbe

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from rudycity/superagent

AGENTS.md@.agents AGENTS.md file
rudycity/superagent · .agents/AGENTS.md
git:20261015.af82465 · audit B · 52 stars
accessibility-compliance SKILL.md skill
rudycity/superagent · .agents/skills/accessibility-compliance/SKILL.md · Implement WCAG 2.2 compliant interfaces with mobile accessibility, inclusive design patterns, and assistive technology…
git:20260626.019dba9 · audit A · 52 stars
adhd SKILL.md skill
rudycity/superagent · .agents/skills/adhd/SKILL.md · Designed to assist users with ADHD (Attention Deficit Hyperactivity Disorder) by mitigating task paralysis, managing…
git:20260724.a9e7b76 · audit A · 52 stars
agent-browser SKILL.md skill
rudycity/superagent · .agents/skills/agent-browser/SKILL.md · Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages…
git:20260626.019dba9 · audit A · 52 stars
airflow-dag-patterns SKILL.md skill
rudycity/superagent · .agents/skills/airflow-dag-patterns/SKILL.md · Build production Apache Airflow DAGs with best practices for operators, sensors, testing, and deployment. Use when…
git:20260626.019dba9 · audit A · 52 stars
algorithmic-art SKILL.md skill
rudycity/superagent · .agents/skills/algorithmic-art/SKILL.md · Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users…
git:20260626.019dba9 · audit B · 52 stars
angular-migration SKILL.md skill
rudycity/superagent · .agents/skills/angular-migration/SKILL.md · Migrate from AngularJS to Angular using hybrid mode, incremental component rewriting, and dependency injection updates…
git:20260626.019dba9 · audit A · 52 stars
anti-reversing-techniques SKILL.md skill
rudycity/superagent · .agents/skills/anti-reversing-techniques/SKILL.md · Understand anti-reversing, obfuscation, and protection techniques encountered during software analysis. Use this skill…
git:20260626.019dba9 · audit A · 52 stars
antislop-code SKILL.md skill
rudycity/superagent · .agents/skills/antislop-code/SKILL.md · Code comment hygiene for AI coding agents: remove generic AI-slop comments, keep the valuable ones, never touch the…
git:20260919.e65d925 · audit A · 52 stars
antislop-copywriting SKILL.md skill
rudycity/superagent · .agents/skills/antislop-copywriting/SKILL.md · Copy and text skill for antislop. Use when writing or editing prose: headlines, tone, CTAs, and anti-AI-writing…
git:20260919.e65d925 · audit A · 52 stars
antislop-human SKILL.md skill
rudycity/superagent · .agents/skills/antislop-human/SKILL.md · Human and accessibility skill for antislop. Contrast, keyboard, focus, and states for real people. Includes the…
git:20260919.e65d925 · audit A · 52 stars
antislop-layoutmobile SKILL.md skill
rudycity/superagent · .agents/skills/antislop-layoutmobile/SKILL.md · Mobile layout skill for antislop. Use for layouts that reflow across screen sizes, phone to desktop: grids, overflow…
git:20260919.e65d925 · audit A · 52 stars

Every file in rudycity/superagent

Other files named karpathy-guidelines

karpathy-guidelines SKILL.md skill
hashgraph-online/awesome-codex-plugins · plugins/Colin4k1024/tsp/skills/karpathy-guidelines/SKILL.md · 提供一组面向编码行为的轻量约束,帮助代理在动手前暴露假设、优先简单方案、保持改动外科化,并把任务改写成可验证目标。 当任务容易被误解、过度设计、顺手扩散改动,或需要先锁定成功标准时使用。
git:20260512.6523792 · audit A · 1,352 stars
karpathy-guidelines SKILL.md skill
coco-research/coco · skills/karpathy-guidelines/SKILL.md · Pointer to the upstream skill for behavioural guidelines that reduce common LLM coding mistakes, installed from…
git:20261008.3d24d40 · audit A · 827 stars
karpathy-guidelines SKILL.md skill
mxyhi/ok-skills · karpathy-guidelines/SKILL.md · Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid…
git:20260508.0cab7e8 · audit A · 504 stars
karpathy-guidelines SKILL.md skill
cowork-os/cowork-os · resources/skills/karpathy-guidelines/SKILL.md · Surgical execution guardrails for coding, debugging, review, and refactor tasks
v1.0.0 · audit A · 496 stars
karpathy-guidelines SKILL.md skill
openyida/openyida · .agents/skills/karpathy-guidelines/SKILL.md · Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid…
git:20260728.2c22517 · audit A · 228 stars
karpathy-guidelines SKILL.md skill
index-login/mobilere-skill · .kilo/skill/karpathy-guidelines/SKILL.md · 减少 LLM 常见编码错误的行为准则。在编写、审查或重构代码时使用,避免过度设计、精准修改、暴露假设、定义可验证的成功标准。
git:20260803.71f8be9 · audit A · 203 stars
karpathy-guidelines SKILL.md skill
youdotcom-oss/agent-skills · .agents/skills/karpathy-guidelines/SKILL.md · Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid…
git:20260723.e470633 · audit A · 93 stars
karpathy-guidelines SKILL.md skill
jjmartres/ai-coding-agents · shared/.ai-agents/skills/karpathy-guidelines/SKILL.md · Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid…
git:20260428.dfccef3 · audit A · 48 stars

Browse by kind, by grade A, or by owner.