Andrej Karpathy's LLM Coding Skills
Single
CLAUDE.mdguardrails derived from Karpathy's public observations on LLM coding failures; repomultica-ai/andrej-karpathy-skillshit ~185K GitHub stars after viral 2026 threads, including a Spanish-language April thread framing discipline as the coding-agent bottleneck.
What it is
Community-encoded principles in one file (also Claude Code plugin andrej-karpathy-skills@karpathy-skills). Addresses silent wrong assumptions, overcomplicated abstractions, orthogonal edits, and weak success criteria. Source: github.com/multica-ai/andrej-karpathy-skills, 2026-06-01
As of 2026-07-01, the repo is on main at 2c606141936f1eeef17fa3043a72095b4765b9c2, with no GitHub release tags and no license field exposed by the GitHub API; GitHub reported 186,021 stars and 19,088 forks. Treat it as a high-signal instruction source, not a package with semantic version authority. Source: GitHub API; git ls-remote, 2026-07-01
Four principles
| Principle | Addresses |
|---|---|
| Think Before Coding | Wrong assumptions, hidden confusion, missing tradeoffs |
| Simplicity First | Bloat, speculative features, 200 lines when 50 suffice |
| Surgical Changes | Drive-by refactors, touching unrelated code |
| Goal-Driven Execution | Tests-first loops, verifiable checkpoints |
Source: github.com/multica-ai/andrej-karpathy-skills, 2026-06-01
Karpathy insight quoted upstream: give success criteria and let the model loop, not imperative micromanagement. Source: github.com/multica-ai/andrej-karpathy-skills README, 2026-06-01
Spanish thread context
Nico Bilinkis thread (527 likes, 2,196 bookmarks): one CLAUDE.md, no CLI, no dependencies, paste at repo root. Stressed discipline over model capacity as the 2026 coding-agent bottleneck. Compared hype around harnesses and MCPs to a file that subtracted instead of added. The thread has no local visual media; the reviewed artifacts were the resolved repo README and the self-thread. Source: X/@nicobilinkis, 2026-04-14
Repo stars at enrichment: 163,925 (prior wiki entry cited ~3.7K from an earlier @sharbel signal); current source check: 186,021. Source: github.com/multica-ai/andrej-karpathy-skills, 2026-06-01; GitHub API, 2026-07-01
Sharbel launch artifact
The @sharbel bookmark is the earliest reviewed screenshot in this cluster. It shows the README's "Karpathy-Inspired Claude Code Guidelines" page and maps Karpathy's failure observations to four principles: Think Before Coding, Simplicity First, Surgical Changes, and Goal-Driven Execution. The screenshot is useful because it makes the mapping visual and teachable; it does not change the local adoption decision. Source: X/@sharbel visual artifact, 2026-04-11
Relevance
Same pattern as Kevin's skills taxonomy: upstream expertise becomes executable config. Pairs with Claude Code Harness and superpowers brainstorming/TDD skills.
Local adoption boundary: do not import this as a competing universal rule file. Kevin already owns the same invariants through Agent Ethos, Code Taste, Browser Testing Skills, Security and Review Skills, and the global AGENTS contract: state assumptions, prefer simple code, make surgical edits, and define verifiable success criteria. The repo remains useful as external evidence and a compact teaching artifact; local behavior should stay in Kevin-owned skills and rules. Source: X/@meta_alchemist, 2026-06-09
Timeline
- 2026-06-30 | Deep-reviewed the @nicobilinkis Spanish thread, self-thread, and resolved GitHub README. Decision: no new skill, because the four rules are already covered by Kevin-owned
AGENTS.md, Agent Ethos, Code Taste, Browser Testing Skills, and Security and Review Skills. Versioning decision: no public package/release version; pin source tomainHEAD2c606141936f1eeef17fa3043a72095b4765b9c2. Source: X/@nicobilinkis, 2026-04-14; Source: GitHubmultica-ai/andrej-karpathy-skills, 2026-06-30 - 2026-06-30 | Deep-reviewed the Meta Alchemist bookmark and repaired the local visual artifact. Current source snapshot:
mainHEAD2c606141936f1eeef17fa3043a72095b4765b9c2, no releases/tags, no license field, 185,077 stars. Promoted as source evidence for Kevin's existing assumptions/simplicity/surgical-change/success-criteria rules, not as a duplicate local skill. Source: X/@meta_alchemist, 2026-06-09; Source: GitHubmultica-ai/andrej-karpathy-skills, 2026-06-30 - 2026-07-01 | Deep-reviewed the @sharbel artifact and refreshed GitHub metadata. Current source snapshot: unchanged HEAD
2c606141936f1eeef17fa3043a72095b4765b9c2, no tags, no license field, 186,021 stars, 19,088 forks. Decision unchanged: reference-only because Kevin's local rules already own these invariants. Source: X/@sharbel, 2026-04-11; GitHub, 2026-07-01 - 2026-04-11 | @sharbel signal (~3.7K stars that week). Source: X/@sharbel, 2026-04-11
- 2026-04-14 | Spanish viral thread by @nicobilinkis; star count exploded. Source: X/@nicobilinkis, 2026-04-14
- 2026-06-01 | Stars and thread context updated from enriched bookmark
2044112901720481986. Source: compiled from enrichment, 2026-06-01