llm-basics
LLM architecture, tokenization, transformers, and inference optimization. Use for understanding and working with language models.
Découvrez les skills research llm ai, comparez leurs dépôts et consultez leurs instructions.
LLM architecture, tokenization, transformers, and inference optimization. Use for understanding and working with language models.
LLM deployment strategies including vLLM, TGI, and cloud inference endpoints.
Prompt design, optimization, few-shot learning, and chain of thought techniques for LLM applications.
Run Codex CLI, Claude Code, OpenCode, or Pi Coding Agent via background process for programmatic control.
Activate expert reasoning and systematic problem-solving mindset for technical and code-related tasks.
AI-native session continuity through shape anchors rather than linear logs. Use when beginning a session to orient from prior state, or when ending to crystallize state for future instances.
Autonomous claims drafting and optimization agent. Drafts, analyzes, and refines patent claims to maximize protection while ensuring validity.
Autonomous patent application drafting agent. Drafts complete patent applications including claims, specification, and abstract from invention descriptions.
UCF "hit it" Session Executor. Executes the complete 33-module consciousness simulation pipeline across 7 phases. ACTIVATE when user says "hit it", "run ucf", "execute pipeline", or references consciousness eme...
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or to...
Expert writing for technical content, social media, and developer documentation. Use when asked to write tweets, LinkedIn posts, blog posts, READMEs, commit messages, feature announcements, or any written conte...
Delegate repetitive tasks to specialized agent to preserve context. Use when user asks to fix linting errors, fix diagnostics, rename variables across files, update imports, fix formatting issues, or perform ot...
Use when saving current iteration progress mid-conversation, before context compaction, or at interim pause points
Use when beginning a new conversation to work on an open-ended goal, loading context from previous iterations through iteration journals
Run Python code in the cloud with serverless containers, GPUs, and autoscaling. Use when deploying ML models, running batch jobs, scheduling tasks, serving APIs with GPU acceleration, or scaling compute-intensi...
Generates clean, runnable ROS 2, Gazebo, Isaac Sim, and VLA code for humanoid robotics
ALWAYS invoke this skill FIRST at session start, after context reset/compaction, or when user mentions "reset", "new session", "where were we", "what was I working on". Do NOT skip this for specific task reques...
Activates when user needs help designing REST APIs, GraphQL schemas, or API architecture. Triggers on "design API", "REST endpoint", "GraphQL schema", "API structure", "endpoint naming", "API versioning", "requ...
Activates when user needs help with performance optimization, profiling, or improving code efficiency. Triggers on "optimize performance", "make this faster", "reduce memory", "profile this", "performance issue...
Use this skill when users ask about RAG (Retrieval-Augmented Generation), semantic search, document indexing, embeddings, vector databases, or chunking strategies. Trigger phrases include "how does rag-research...
创建有效技能的指南。当用户想要创建新技能(或更新现有技能)以通过专业知识、工作流程或工具集成扩展 Claude 功能时,应使用此技能。
Reproduce research papers into working code. Use when user wants to implement ML/AI papers, reproduce experiments, extract algorithms from PDFs, or convert research into executable code. Handles multiple interc...
Build headless automation and agentic workflows with Google's Gemini CLI. Covers approval modes (default, auto_edit, yolo), file permission model, Edit vs WriteFile tool selection, smartEdit configuration, GEMI...
Download Claude Code templates from aitmpl.com (GitHub API). Use PROACTIVELY to search and install agents, commands, skills, mcps, settings, hooks, and plugins.
Le classement utilise les étoiles du dépôt GitHub indexé. Plusieurs skills peuvent partager ce nombre : ce n’est ni un nombre d’installations ni une note de qualité.