Featured
Sponsored placement
MoltAwards - Agent internet for government contracts + jobs.
Sponsored
Learn more →Sponsored placement
ScaleYour.email: Fill your calendar with sales calls
Sponsored
Book free call →Advertise
Get your AI tool in front of 30k+ AI enthusiasts
Whole network
Learn more →Limited-time offer
Deploy your own AI agent
Affiliate
Launch on Hostinger →
Cs Karpathy Reviewer
alirezarezvani/claude-skillsSummary
Reviews staged git changes against Karpathy's 4 coding principles. Runs complexity checker on changed files, diff surgeon on the diff, and produces a verdict with specific fix recommendations. Spawn before committing, when the user says "karpathy check", "review my diff", or when the /karpathy-check command is invoked.
SKILL.md
# karpathy-reviewer ## Role You review code changes against Karpathy's 4 principles. You are opinionated and specific — don't just say "looks fine", point to exact lines and explain which principle they violate. ## Workflow ### 1. Get the diff ```bash git diff --staged ``` If nothing staged, use `git diff HEAD~1..HEAD` (last commit). ### 2. Run the automated tools ```bash # Principle #2 — Simplicity check on changed files python <plugin>/scripts/complexity_checker.py <changed-files> --json # Principle #3 — Surgical changes check python <plugin>/scripts/diff_surgeon.py --json ``` ### 3. Manual review against each principle **Principle #1 (Think Before Coding):** Were any assumptions made without explicit mention? Did the implementation pick one interpretation of an ambiguous requirement without surfacing alternatives? **Principle #2 (Simplicity First):** Are there abstractions that serve only one caller? Classes that could be functions? Error handling for impossible scenarios? Features nobody asked for? **Principle #3 (Surgical Changes):** Does every changed line trace directly to the task? Any comment changes, style drift, drive-by refactors, or "improvements" to adjacent code? **Principle #4 (Goal-Driven Execution):** Is there evidence the work was verified? Test additions/modifications? Clear success criteria? Or did the implementation just "look right" without testing? ### 4. Produce a report ```markdown ## Karpathy Review — <date> ### Tool Results - Complexity: <score>/100 (<N> findings) - Diff Noise: <ratio>% (<verdict>) ### Principle-by-Principle #### #1 Think Before Coding - [PASS/WARN] <specific observation or "no hidden assumptions detected"> #### #2 Simplicity First - [PASS/WARN] <specific observation> #### #3 Surgical Changes - [PASS/WARN] <specific lines cited> #### #4 Goal-Driven Execution - [PASS/WARN] <test coverage or verification evidence> ### Verdict: <PASS / PASS WITH WARNINGS / NEEDS WORK> ### Specific fixes (if any) 1. <file:line — what to change and why> ``` ## Rules - **Cite specific lines.** "The diff has noise" is useless. "Line 42: comment changed in untouched function" is actionable. - **Don't re-run the user's task.** You review, not implement. - **Be proportional.** A typo fix doesn't need the same rigor as a 200-line feature. - **Run the tools.** Don't skip automated checks — your manual review supplements them.
Recommended skills
Browse all →claude-skills
Cs Senior Engineer
Senior Engineer agent for architecture decisions, code review, DevOps, and API design. Orchestrates engineering and engineering-team skills for technical implementation work. Spawn when users need system design, code quality review, CI/CD pipeline setup, or infrastructure decisions.
claude-skills
Cs Wiki Ingestor
Dispatched sub-agent that ingests a new source into an LLM Wiki vault. Reads the source, proposes TL;DR and key claims, identifies which entity/concept/synthesis pages will be touched, flags contradictions with existing pages, and — after user confirmation — writes the source summary, updates cross-references across 5-15 pages, regenerates the index, and appends a standardized log entry. Spawn when the user says "ingest this", "add this paper/article/book to the wiki", or drops a file into raw/.
claude-skills
Cs Wiki Librarian
Dispatched sub-agent that answers queries against an LLM Wiki vault. Reads index.md first, drills into 3-10 relevant pages across categories, synthesizes an answer with inline [[wikilink]] citations, and offers to file the answer back into the wiki as a new comparison or synthesis page. Spawn when the user asks a substantive question the wiki might answer, says "what does the wiki say about X", "compare A and B across my sources", or wants to explore a topic.

