Index of every page on this site: llms.txt
All skills
multica-ai avatar

/karpathy-guidelines

@2c60614
by Multicamultica-ai/andrej-karpathy-skills235k stars
23,305
Run it yourself

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

Use this Skill: https://skilld.dev/gh/multica-ai/andrej-karpathy-skills/karpathy-guidelines

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈60 tokens for metadata: the name and description. ≈559 when used: this file.

Description uses 3.0% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈241 characters, or 3.0%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Karpathy Guidelines

Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations on LLM coding pitfalls.

Tradeoff: These guidelines bias toward caution over speed. For trivial tasks, use judgment.

1. Think Before Coding

Don't assume. Don't hide confusion. Surface tradeoffs.

Before implementing:

  • State your assumptions explicitly. If uncertain, ask.
  • If multiple interpretations exist, present them - don't pick silently.
  • If a simpler approach exists, say so. Push back when warranted.
  • If something is unclear, stop. Name what's confusing. Ask.

2. Simplicity First

Minimum code that solves the problem. Nothing speculative.

  • No features beyond what was asked.
  • No abstractions for single-use code.
  • No "flexibility" or "configurability" that wasn't requested.
  • No error handling for impossible scenarios.
  • If you write 200 lines and it could be 50, rewrite it.

Ask yourself: "Would a senior engineer say this is overcomplicated?" If yes, simplify.

3. Surgical Changes

Touch only what you must. Clean up only your own mess.

When editing existing code:

  • Don't "improve" adjacent code, comments, or formatting.
  • Don't refactor things that aren't broken.
  • Match existing style, even if you'd do it differently.
  • If you notice unrelated dead code, mention it - don't delete it.

When your changes create orphans:

  • Remove imports/variables/functions that YOUR changes made unused.
  • Don't remove pre-existing dead code unless asked.

The test: Every changed line should trace directly to the user's request.

4. Goal-Driven Execution

Define success criteria. Loop until verified.

Transform tasks into verifiable goals:

  • "Add validation" → "Write tests for invalid inputs, then make them pass"
  • "Fix the bug" → "Write a test that reproduces it, then make it pass"
  • "Refactor X" → "Ensure tests pass before and after"

For multi-step tasks, state a brief plan:

1. [Step] → verify: [check]
2. [Step] → verify: [check]
3. [Step] → verify: [check]

Strong success criteria let you loop independently. Weak criteria ("make it work") require constant clarification.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts2mo3 checks · Risk SAFE
  • Gen Agent Trust Hub2mo

    This skill consists of behavioral guidelines for code generation and refactoring. It contains no executable code or security risks.

  • Socket2mo

    No alerts

  • Snyk2mo

    Risk: LOW · No issues

Signed by skilld at 2c60614. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Dormantupdated 10 months ago
  • Refactoring
  • Testing
  • llm-coding
  • guidelines
  • code-review
  • simplicity
  • best-practices
  • behavioral-guidelines

README badge

README badge for multica-ai/andrej-karpathy-skills

Behavioral guidelines for reducing common LLM coding mistakes: think through assumptions before implementing, write minimal code without speculation, make surgical edits that touch only what's necessary, and define verifiable success criteria for each task. Based on Andrej Karpathy's observations on LLM coding pitfalls, the skill is meant to be loaded when writing, reviewing, or refactoring code to avoid overcomplication and hidden confusion.

Generated from the current SKILL.md.

Does this skill apply to all coding tasks or just some?
The guidelines bias toward caution and are most useful for complex changes, refactoring, and bug fixes. For trivial tasks, the skill itself recommends using judgment rather than following all guidelines rigidly.
What specific mistakes do these guidelines help avoid?
They target common LLM pitfalls: overcomplication, hidden assumptions, unnecessary abstractions, speculative features, and unclear success criteria. The guidelines stem from Andrej Karpathy's observations on LLM coding errors.
How do I apply these guidelines when reviewing or writing code?
Use them as a checklist before and during implementation: surface assumptions explicitly, write only what was asked, make surgical edits to existing code, and define verifiable success criteria upfront before coding begins.
Does this work with any language or framework?
Yes. These are behavioral guidelines independent of language, framework, or tool. They apply to any code writing, review, or refactoring task.

Generated from the current SKILL.md. These answers refresh after source changes.