Découvrez les skills Tests et sécurité, comparez leurs dépôts et consultez leurs instructions.
Enable Nobel Prize and Fields Medal level mathematical reasoning and breakthrough capabilities in SO8T models through alpha gate sigmoid control, golden ratio phi^(-2) convergence, and induced grokking phenomen...
Develop comprehensive evolution plan for Qwen2.5-7B to SO8T/thinking model with advanced Japanese capabilities, mathematical reasoning at Nobel/Fields medal level, and integration of 2024-2026 LLM breakthroughs...
Integrate DeepSeek-R1 GRPO, mHC manifold constraints, geometric scaling, and SO8T quadrality inference into Sunset Pipeline for advanced LLM evolution. Use when implementing cutting-edge mathematical reasoning...
Create comprehensive implementation plans for integrating GRPO, mHC manifold constraints, geometric scaling, SO8T quadrality inference, and imatrix quantization into Sunset Pipeline for advanced LLM evolution....
Capture learnings from current session for future knowledge compounding
Creates new AI agent skills following the Agent Skills spec. Trigger: When user asks to create a new skill, add agent instructions, or document patterns for AI.
This skill should be used when the user asks to "resume session", "save state", "restore context", "view session history", "create checkpoint", or when the /resume command is invoked. Provides session state tra...
Expert system for Ultralytics YOLOv8/v9/v10. Provides documentation on training, detection, segmentation, and deployment.
Query CYNIC's philosophical wisdom across 19 domains and 73 engines. Use when asked philosophical questions, need cross-domain synthesis, want philosophical grounding for decisions, or seek wisdom from multiple...
Acts as a Socratic Product Manager to explore, clarify, and define new application ideas. Use when the user has a vague idea, wants to "brainstorm", or asks to "design this app".
LUTAGU 技能索引。當用戶提到專案開發、介面設計、交通票價、 醫療查詢等主題時，作為一級路由引導 AI 載入具體 Skill。
Use when integrating Context7 (knowledge/context store) for document ingestion, semantic search, or scoped context retrieval. Triggers for: uploading documents, searching knowledge base, filtering by role/tenan...
Understand the components, mechanics, and constraints of context in agent systems. Use when designing agent architectures, debugging context-related failures, or optimizing context usage.
Gerador e executor de quiz sobre conhecimento RAG. Comandos: '/gerar-quiz-rag gerar' para gerar perguntas do banco RAG e salvar em JSON, '/gerar-quiz-rag iniciar' para executar quiz de 10 perguntas, '/gerar-qui...
Use this skill when the user asks to "test prompts", "create prompt tests", "evaluate prompt quality", "run regression tests on prompts", "compare prompt versions", "set up LLM-as-judge evaluation", or needs gu...
Create or update AgentSkills. Use when designing, structuring, or packaging skills with scripts, references, and assets.
Generative Engine Optimization for AI search engines (ChatGPT, Claude, Perplexity).
Comprehensive expert knowledge for NVIDIA DGX Spark workstation. Use when users ask about DGX Spark hardware, software, playbooks, AI Workbench, fine-tuning, inference, troubleshooting, known issues, container...
Use when starting any conversation - establishes how to find and use skills before ANY response including clarifying questions
Determine consensus/disagreement from 5-agent debate - exposes disagreements as developer decisions
Agent Behavior System governance layer for AI coding agents - enforce policies, audit decisions, and maintain safe autonomy
Run comprehensive AI and recommendation system tests. Creates test users at different game stages (Gen 10 dolphin, Gen 4 F2P, Gen 2 whale) with farm accounts, tests recommendations, lineups, and AI question rou...
Field Meta-Orchestrator. Compiles and instantiates the appropriate skills for every perturbation (Unified Aethelred + COAC).
Use when asked to regenerate audio examples with shorthand like "L15 > P1 > E2" or "lesson 15 pattern 1 example 2" - runs regenerate_example.py with correct parameters
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.