LLM & Agents

Facilitates context sharing and strategic delegation between Claude Code and Gemini CLI. Syncs CLAUDE.md to GEMINI.md and provides agent selection guidance. Use when onboarding Gemini to a project, syncing instructions between agents, or deciding whether to use Claude or Gemini for a specific task.

This skill provides the Gen Alpha/brainrot output style transformation rules and vocabulary. It is automatically loaded by the SessionStart hook to transform all Claude responses into Gen Alpha internet slang. Contains comprehensive glossary, transformation patterns, and examples.

Design custom agents from scratch using Claude Agent SDK patterns. Use when building domain-specific agents, designing agents with full SDK control, or creating specialized agents with custom tools and prompts.

Guide for creating custom Claude Code slash commands with proper structure, argument handling, frontmatter configuration, and best practices. Use when the user wants to create slash commands, custom commands, reusable prompts, or mentions creating/designing/building commands.

Leverage AI tools effectively for professional writing tasks. Provides prompt patterns, refinement workflows, voice preservation techniques, and quality checkpoints for AI-augmented drafting.

ベクトルデータベースとセマンティック検索を使用してLLMアプリケーションのためのRetrieval-Augmented Generation（RAG）システムを構築します。知識に基づいたAI、ドキュメントQ&Aシステムの構築、またはLLMと外部ナレッジベースの統合時に使用します。

Creates long-running autonomous agents. Use when the user asks for "롱 러닝 에이전트 만들어줘", "자율 에이전트 생성", "autonomous agent", "long-running agent", "nonstop agent", or "24/7 agent". Collects requirements through AskUserQuestion and generates agent structure following Anthropic best practices.

Generates custom Claude Code subagents with specialized expertise. Activates when user wants to create a subagent, specialized agent, or task-specific AI assistant. Creates properly formatted .md files with YAML frontmatter, suggests tool restrictions and model selection, generates effective system prompts. Use when user mentions "create subagent", "new agent", "specialized agent", "task-specific agent", or wants isolated context for domain-specific work.

Guidance library grammar-constrained generation with categorical structure. Use when implementing structured LLM outputs with grammar constraints, building template-based generation with guaranteed formats, creating type-safe prompt templates with interleaved computation, or applying categorical grammar theory to constrained generation.

Unified four-tier memory system for AI agents. Tier 1 Semantic (Serena+Forgetful search), Tier 2 Episodic (session replay), Tier 3 Causal (decision patterns). Enables memory-first architecture per ADR-007.

Store universal coding patterns and insights into vector database. Use after completing difficult tasks with broadly-applicable lessons that work across ANY project. Use when user says "--coder-store" or "--learn" (you decide which scope, may use both) or after discovering framework-agnostic patterns. ALSO invoke when user expresses strong frustration using trigger words like "fuck", "fucking", "shit", "moron", "idiot", "stupid", "garbage", "useless", "terrible", "wtf", "this is ridiculous", "you're not listening" - these are CRITICAL learning signals for storing failure patterns. Skip for routine work or project-specific details. Auto-invokes based on task completion and user signals.

Manage vim-ai plugin configuration files including .vimrc settings, custom roles, API keys, and model configurations. Use when user requests help with vim-ai setup, modifying vim-ai behavior, adding custom roles, changing AI models, updating prompts, or troubleshooting vim-ai issues.

Expert assistant for calculating materials properties from first-principles using ASE - structure relaxation, surface energies, adsorption, reaction barriers, phonons, elastic constants, and thermodynamic modeling with proper scientific methodology

Comprehensive PRP (Product Requirements Prompt) workflow for feature implementation. Use when user wants to implement complex features, needs a structured implementation plan, or mentions PRPs, feature implementation, or systematic development. Handles PRP generation from requirements, clarification of ambiguities, and execution for implementation. Three modes - Generation (create PRP), Clarification (resolve ambiguities), Execution (implement feature).

Design retrieval-augmented generation pipelines including chunking, embedding, retrieval, and context assembly strategies.

Execute Claude Code autonomously in the background for well-defined tasks, enabling parallel execution of up to 5 concurrent sessions while the user continues other work.

Create and improve TypeScript project documentation using TypeDoc. Use when documenting TypeScript codebases, adding JSDoc comments, configuring TypeDoc, auditing documentation coverage, or generating API reference documentation for TypeScript projects.

Create Level 6 template meta-prompts that generate other prompts. Use when building prompt generators, designing high-leverage meta-prompts, or creating templates that scaffold other prompts.

Vibora is a terminal-first tool for orchestrating AI coding agents across isolated git worktrees. Use this skill when working in a Vibora task worktree or managing tasks.

Expert assistant for conducting remote experiments with Claude-Light - a web-accessible RGB LED and spectral sensor instrument for statistics, regression, optimization, and design of experiments