Behavioral guidelines that reduce common LLM coding mistakes — surface assumptions, keep changes simple and surgical, and define verifiable success criteria. Derived from Andrej Karpathy's observations on LLM coding pitfalls, packaged from the multica-ai skills collection.
Plugin karpathy-guidelines from marketplace jazis-ai-marketplace in jazithedev/ai-marketplace. Commands read from the repo's manifest — 4 days ago.
karpathy-guidelinesBehavioral 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.
Editorial diagram design skill with 44 diagram types — architecture, flowchart, sequence, ER, timeline, Gantt, Sankey and more — as self-contained HTML + SVG, with no shadows and no Mermaid look. By Cathryn Lavery.
1 skill · 6 commands
Skills for understanding a problem before building it — /problem-space:problem-analyst maps stakeholders, pain points, requirements, business rules, success criteria, assumptions and risks; /problem-space:open-questions turns that into a short, natural list of the questions only people can answer; /problem-space:user-story-writer slices it into thin, INVEST-compliant user stories. None suggests solutions. Analyst and story writer by Andrea Laforgia.
3 skills
Create draft pull requests from the commits on the current branch — a single PR, or a stacked chain of child PRs under a collective branch — list the GitHub pull requests awaiting your code review, and board the open PRs you authored with their review, checks, merge state and stack topology. Provider-agnostic creation: full GitHub and GitLab support, honest fallback elsewhere; always previews the full PR plan before creating anything.
3 skills
An always-on tone of voice that keeps everything Claude writes short, plain and easy for non-native English speakers to read, without losing facts, decisions or risks. Applies itself to chat replies, investigations, tickets, PRs and docs; switch it with /say-less:say-less off|on.
1 skill · hooks
Multi-agent code review skill simulating JaziTheDev's review style — PR discipline, bugs and smells, project-rules compliance, historical context, code documentation, and tactical/strategic DDD. Posts findings as inline GitHub review comments. Invoked as /code-review:code-review.
1 skill
Create and update well-structured GitLab work items — Context, Expected Result, numbered full-sentence acceptance criteria, and optional Action Points, QA Notes and Implementation Plan sections. Keeps implementation detail out of the acceptance criteria, retires Action Points as they are answered, and updates existing items safely via the glab CLI.
1 skill
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Harness-native ECC operator layer - 71 agents, 314 skills, 98 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
200 skills · 98 commands · 71 agents · 1 MCP server · hooks
Claude API and SDK documentation skill for building LLM-powered applications
19 skills
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
2 skills · 2 commands · 2 agents · 1 MCP server
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
1 command · 6 agents
Professional UI/UX design intelligence for AI coding assistants. Includes searchable databases of styles, colors, typography, charts, and UX guidelines for React, Next.js, Astro, Vue, Nuxt.js, Nuxt UI, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, Jetpack Compose, Angular, Laravel, JavaFX, and Three.js.
7 skills
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.