karpathy-guidelines
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.
85
80%
Does it follow best practices?
Impact
92%
1.14xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/karpathy-guidelines/SKILL.mdEvaluation results
80%
Simplify the Domain Extraction Utility
Surgical code edits with orphan cleanup
Orphaned import removed
0%
0%
Non-orphaned import kept
100%
100%
Dead function preserved
100%
100%
Unchanged functions untouched
100%
100%
Style matched
100%
100%
Dead code mentioned
100%
100%
No added annotations
100%
100%
100%
Log Error Extractor
Simplicity first, no over-engineering
No class abstraction
100%
100%
No extra CLI flags
100%
100%
No config file
100%
100%
No speculative error handling
100%
100%
No extra output files
100%
100%
No unrequested features
100%
100%
Concise implementation
100%
100%
No logging framework
100%
100%
No type annotations on simple code
100%
100%
98%
37%Extend the User Export Script with Filtering
Goal-driven execution with assumption surfacing
Multiple interpretations listed
33%
100%
Chosen interpretation stated
91%
100%
Reasoning given for choice
83%
100%
Numbered implementation plan
33%
83%
Per-step verification
0%
100%
Simpler path noted
90%
100%
No gold-plating
83%
100%
Existing functions unchanged
100%
100%
- Repository
- forrestchang/andrej-karpathy-skills
- Commit
2c60614
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.