Loading
Loading
written by ForgeCat

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.
npx forgecat install @forgecat/andrej-karpathy-skills
| Field | Value |
|---|---|
| Author | Multica AI |
| Original repository | https://github.com/multica-ai/andrej-karpathy-skills |
| Version | 0.0.9 |
| Original commit | 2c606141936f1eeef17fa3043a72095b4765b9c2 |
| License | MIT |
| Source platform | multi-host |
| Platform | Status |
|---|---|
| Claude Code | Tested |
| Cursor | Tested |
| Codex | Tested |
| OpenClaw | Partial |
| Hermes | Partial |
SKILL.md files.Upstream declares MIT. The original declaration and pinned source are preserved in for-forgecat/LICENSE-SOURCE.md and in the installed skill’s forgecat-notices/ directory. MIT-STANDARD-REFERENCE.md is a ForgeCat-added reference, not an upstream LICENSE. No copyright holder or year has been supplied by ForgeCat. The upstream missing copyright notice remains disclosed.
This is a locally prepared candidate. Fresh exact-version installation and runtime verification are pending on all five platforms; no Tested claim is inherited. External service access, application deployment and model output quality have not been tested.
The skill and both original project rule sources are preserved. OpenClaw/Hermes currently receive the skill without the separate project rules; their candidate scope remains Partial.
forgecat
View CreatorNone
Evaluated 9/22/2026