karpathy-coder

Use when writing, reviewing, or committing code to enforce Karpathy's 4 coding principles — surface assumptions before coding, keep it simple, make surgical changes, define verifiable goals. Triggers on "review my diff", "check complexity", "am I overcomplicating this", "karpathy check", "before I commit", or any code quality concern where the LLM might be overcoding.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

—

The risk profile of this skill

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-organized, lean body that pairs tight principles with concrete tooling and a clear enforcement workflow. The main gaps are minor: slight expansiveness in compatibility/related-skills sections and a referenced hook file absent from the bundle.

Suggestions

Add a hooks/ directory (or note that karpathy-gate.sh is generated by an installer) so the referenced pre-commit hook matches the actual bundle structure.

Trim the 'Cross-tool compatibility' table to a single line or move it to references/enforcement-patterns.md to tighten the overview.

Inline a tiny executable snippet (e.g., a one-line complexity_checker.py invocation with expected output) so the actionability anchor for copy-paste-ready examples is fully met.

Dimension	Reasoning	Score
Conciseness	Lean and efficient with tight bullets and no padding of concepts Claude already knows; minor trimming possible in the 'Cross-tool compatibility' and 'Related skills' sections, keeping it just below the lean-5 anchor.	4 / 5
Actionability	Provides concrete executable guidance — the 'Instead of... Transform to...' table, plan template, script-to-detection table, and slash command — with minor gaps since the script implementations live in the bundle rather than inline.	4 / 5
Workflow Clarity	Clear sequence via the pre-commit hook and /karpathy-check dispatching the reviewer agent, plus the Goal-Driven plan-with-verify template encoding a feedback loop; not a 5 because the skill is principle-driven rather than a fragile multi-step operation with explicit validation checkpoints.	4 / 5
Progressive Disclosure	Well-signaled one-level-deep references (all three reference files and four scripts exist and are described), with the body kept as an overview; held below 5 by the inlined cross-tool table and a referenced hooks/karpathy-gate.sh that has no corresponding hooks bundle directory.	4 / 5
	Total	16 / 20 Passed

Description

91%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, well-structured description that explicitly states both capability and trigger conditions with natural user phrases. The main weakness is mild overlap with broader code-review skills and principles-versus-actions framing.

Dimension	Reasoning	Score
Specificity	Names the domain and enumerates the four concrete principles ('surface assumptions before coding, keep it simple, make surgical changes, define verifiable goals'), giving specific actions; falls short of 5 because the listed items are principles rather than discrete, comprehensive operations.	4 / 5
Completeness	Explicitly answers both 'what' (enforce the 4 listed principles) and 'when' ('Use when writing, reviewing, or committing code' plus concrete trigger phrases), matching the top anchor.	5 / 5
Trigger Term Quality	Comprehensive coverage of natural user phrases including synonyms and the eponymous trigger: 'review my diff', 'check complexity', 'am I overcomplicating this', 'karpathy check', 'before I commit'.	5 / 5
Distinctiveness Conflict Risk	Clear niche (Karpathy's LLM-specific pitfalls) with a distinctive eponymous trigger ('karpathy check') and minimal conflict risk; not a 5 because 'code quality concern' overlaps with generic code-review skills.	4 / 5
	Total	18 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: alirezarezvani/claude-skills
Path: engineering/karpathy-coder/skills/karpathy-coder/SKILL.md
Commit: 19392f7

Reviewed: ~5 hours ago

Table of Contents

Discovery Implementation Validation

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.