conversation
Bidirectional voice conversation with Push-to-Talk. Use when user says: 'conversation mode', 'let's talk', 'parlons', 'voice conversation', 'dialogue vocal', 'PTT mode', or wants to speak WITH Claude (not just listen). For one-way TTS (Claude speaks, user types), use /speak instead.
platform-integration-workflow
Guide for adding new AI platform support (e.g., Gemini, Mistral, Anthropic) to the Chrome extension with strategy pattern and testing requirements
skill-writer
Use this skill to design, document, and structure AI agent skills for Gemini, Claude, and Codex. It provides architectural rules, directory standards, and writing best practices.
llm-evaluation
自動メトリクス、人間によるフィードバック、およびベンチマークを使用して、LLMアプリケーションの包括的な評価戦略を実装します。LLMのパフォーマンスをテストし、AIアプリケーションの品質を測定し、または評価フレームワークを確立する場合に使用します。
skills-registry
Dynamic skill discovery and matching for any task context. Invoke when: - Starting any task to discover relevant skills - Need to find specialized skills for current context - Agent/command needs skill recommendations Trigger phrases: "match skills", "find skills", "what skills for", "recommend skills", "which capabilities", "skill lookup", "applicable skills" Examples: - "Match skills for TDD cycle with zod validation" → outputs backend-zod, backend-vitest - "What skills for React landing page?" → outputs frontend-master, frontend-aceternity
slash-command
Create and edit Claude Code slash commands. Use when the user wants to create a new slash command, add a custom command, or modify an existing command in .claude/commands/.
foundational-principles
基础原则 - AI 助手的核心思维原则和指令框架概述。包含系统思维、辩证思维、创新思维和批判思维四大核心原则。
nnsight-basics
Core nnsight concepts for neural network interpretability. Use when setting up models, tracing activations, saving values, or making basic interventions on model internals.
cursor-agents-md
Guide for creating effective AGENTS.md files for Cursor. Use when (1) Creating new AGENTS.md files in project root, (2) Improving existing agent instructions, (3) Setting up project-specific coding guidelines, (4) Defining code style and architecture rules, (5) Writing simple agent instruction files as alternative to .cursor/rules, or any task related to Cursor AGENTS.md configuration.
debate-orchestrator
Orchestrates formal debates with proposition and opposition sides, coordinating debaters and judges through structured exchanges. Use when running debate exchanges, managing debate rounds, or continuing interrupted debates.
qwen-ollama
Using Qwen 2.5 models via Ollama for local LLM inference, text analysis, and AI-powered automation
prompt-engineering-patterns
Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production. Use when optimizing prompts, improving LLM outputs, or designing production prompt templates.
linear-algebra-expert
Expert in vector spaces, matrices, linear transformations, eigenvalues, and applications to data science and machine learning
autogluon-tabularpredictor-save
Save TabularPredictor artifacts with TabularPredictor.save, detailing all arguments and persistence behavior; depends on autogluon-tabularpredictor-fit and pairs with autogluon-tabularpredictor-load.
ml-model-explainer
Explain ML model predictions using SHAP values, feature importance, and decision paths with visualizations.
autogluon-tabularpredictor-set-model-best
Set the default prediction model with TabularPredictor.set_model_best, detailing all arguments and effects; depends on autogluon-tabularpredictor-fit and is often guided by autogluon-tabularpredictor-fit-summary or leaderboard outputs.