karpathy-guidelines
Behavioral guidelines for LLMs to reduce coding mistakes, follow best practices, and improve output quality by enforcing simplicity, surgical changes, and goal-driven verification.
Discover reusable agent skills, browse implementation details, and find the right skill for your workflow.
490 skills found
Behavioral guidelines for LLMs to reduce coding mistakes, follow best practices, and improve output quality by enforcing simplicity, surgical changes, and goal-driven verification.
Intelligent pattern selection for Fabric CLI, automatically choosing from 242+ specialized prompts for threat modeling, data analysis, summarization, and content creation.
Implement LlamaExtract for robust structured data extraction from PDF, DOCX, and PPTX files using Pydantic schemas.
Master iOS Human Interface Guidelines and SwiftUI for native app development. Expert guidance for UI design, component implementation, and Apple platform design principles.
Systematically trace code flows, locate implementations, diagnose performance issues, and map system architecture to understand complex codebases.
Automate the migration of Netflix Conductor workflows to Temporal Python, including server orchestration, worker management, and workflow troubleshooting.
Download Instagram Reels via sssinstagram.com and process them into WhatsApp-ready video files automatically.
Python coding assistant providing best practices, PEP 8 enforcement, automated testing with pytest, and dependency management using uv.
Build performant Next.js applications using App Router, Server Components, Server Actions, and modern React best practices.
Query Microsoft 365 Copilot for workplace intelligence—emails, meetings, documents, and team communication—to ground your AI agent in organizational context.
Migrate your codebase, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to the advanced Opus 4.5 model with automated configuration adjustments.
Automated pipeline to download, split, and deeply analyze academic PDFs in structured batches to avoid context window limits and ensure high-quality comprehension.