LLM & Agents

Synchronize documentation with codebase - updates AUTO-GEN sections in CLAUDE.md, validates cross-references, and reports stale documentation

Slidev presentation about taming dependency chaos for LLM in Kubernetes.

LLM-as-judge evaluation patterns, rubric generation, and quality assessment frameworks. Use when evaluating code quality, assessing outputs, creating evaluation criteria, or implementing review systems.

Master advanced AgentDB features including QUIC synchronization, multi-database management, custom distance metrics, hybrid search, and distributed systems integration. Use when building distributed AI systems, multi-agent coordination, or advanced vector search applications.

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 the agent's capabilities with specialized knowledge, workflows, or tool integrations.

Define database models and schemas with proper data types, constraints, relationships, and validation rules for PostgreSQL (Supabase/Bun.sql) and Firestore (Firebase). Use this skill when creating or modifying database models, ORM entity definitions, Prisma schemas, or Firestore document structures. Apply when working on model files (models/*.ts, entities/*.ts, schema.prisma, models/*.py, Models/*.cs), defining database relationships, setting up validation rules, or implementing data integrity constraints. This skill ensures snake_case naming for SQL and camelCase for NoSQL, required timestamps (created_at/updated_at), UUIDs for SQL and auto-generated IDs for Firestore, foreign key constraints with indexed columns, Row Level Security (RLS) policies for Supabase, strict Firestore security rules, normalized data for SQL (3NF) with denormalization for Firestore read performance, and pgvector setup for AI embeddings.

Analyzes project structure and generates appropriate Claude Code subagents based on detected tech stack

Expert Rust code review agent that analyzes Rust code for correctness, safety, idiomatic patterns, performance, and best practices. Use when reviewing Rust code files (.rs), providing feedback on Rust projects, checking for unsafe code patterns, evaluating borrow checker compliance, suggesting idiomatic improvements, or conducting thorough code quality assessments for Rust codebases.

Validate generated experimental procedures against LA-Bench evaluation criteria. This skill orchestrates formal validation (via script) and semantic evaluation (via subagent) to ensure compliance with LA-Bench standards. Use this skill after generating procedures to assess quality before submission.

Interactive slash command creator for Claude Code. Triggers when user mentions creating commands, slash commands, command templates, command arguments, or wants to build a new command workflow.

Automatically updates CLAUDE.md and related documentation when new features, configurations, or best practices are discovered during development sessions

Execute work orders and implement code using coding agents and development tools

Build coding agents with the Claude Agent SDK. Use when creating agentic workflows, implementing todo tracking, monitoring agent progress, building multi-turn conversations, streaming responses, or integrating custom tools with Claude Code.

Plans and executes data migrations between systems, databases, and formats

Chrome Built-in AI APIs for client-side AI features. Use for Writer API, Rewriter API, Proofreader API, Prompt API, Translator API, Language Detector API, Summarizer API, and building AI-powered web applications with on-device processing.

セッションの会話から恒久的な振る舞いルールを抽出し、AGENTS.md と突き合わせて追加すべきものを提示する。出力は rules と agents_md_patch の二部構成。

Manages a structured knowledge base of patterns, troubleshooting guides, best practices, and workflows. Activates when discovering reusable insights, solving technical problems, or establishing new standards. Records knowledge in categorized files for future reference without bloating global CLAUDE.md.

Generate five brand-aligned design concepts for validated niches using Claude creativity constrained by deterministic structure and brand voice memories.

Comprehensive Type 2 executable skill for automated WCS to Godot migration orchestration. Implements Plan-Implement-Validate lifecycle, agent lifecycle management, task state machines, and event-driven hooks with structured JSON I/O for reliable automated remediation loops. Use for coordinating complex migration workflows with deterministic execution.

Validates command frontmatter, delegation patterns, simplicity guidelines, and documentation proportionality. Use when reviewing, auditing, analyzing, evaluating, improving, or fixing commands, validating official frontmatter (description, argument-hint, allowed-tools, model), checking delegation clarity or standalone prompts, assessing simplicity guidelines (6-15 simple, 25-80 documented), validating argument handling, or assessing documentation appropriateness. Distinguishes official Anthropic requirements from custom best practices. Also triggers when user asks about command best practices, whether a command should be a skill instead, or needs help with command structure.