LLM & Agents

Validate Claude Code skills, commands, agents, hooks, and plugins against specifications.Checks structure, format, best practices, and common issues before committing or distributing.Use when: validating skills, checking plugins, linting components, ensuring quality.Activates for: "validate skill", "check skill", "validate plugin", "lint skill", "verify skill"

Use when building a production-ready REST API from requirements through deployment.Orchestrates 8-12 specialist agents across 5 phases using Test-Driven Development methodology.Covers planning, architecture, TDD implementation, comprehensive testing, documentation, andblue-green deployment over a 2-week timeline with emphasis on quality and reliability.

Build OpenCode plugins, agents, hooks, and tools. Use when creating, reviewing, or debugging OpenCode extensions. Covers plugin architecture, agent configuration, lifecycle hooks, custom tools, and distribution. Invoke PROACTIVELY when user mentions OpenCode plugins, agents, hooks, or wants to extend OpenCode functionality.

Create CrewAI workflows as NestJS applications under apps/crewai/. FUTURE: Same pattern as LangGraph - NestJS app, same webhook pattern as n8n, receive same parameters, wrap as API agents. CRITICAL: Status webhook URL must read from environment variables.

Fast code generation agent optimized for speed and iteration

Extracts all chunks with three-tier contextual embeddings from a specific PDF file in Qdrant vector database and saves to plain text. Use when user wants to extract, export, dump, or view all chunks from a PDF document, inspect file content, save chunks for analysis, or review contextual embeddings.

Prompt engineering standards and context engineering principles for AI agents based on Anthropic best practices. Covers clarity, structure, progressive discovery, and optimization for signal-to-noise ratio.

MoAI-ADK's foundational principles - TRUST 5, SPEC-First TDD, delegation patterns, token optimization, progressive disclosure, modular architecture, agent catalog, command reference, and execution rules for building AI-powered development workflows

Manage storage devices and resources in Proxmox. Monitor storage usage, allocate resources, and plan storage expansion for your virtualization infrastructure.

Use when executing implementation plans with independent tasks in the current session or facing 3+ independent issues that can be investigated without shared state or dependencies - dispatches fresh subagent for each task with code review between tasks, enabling fast iteration with quality gates

MUST INVOKE this skill when creating focused prompts in .prompts/prompts/ and multi-stage chains in .prompts/metaprompt/. Create, audit, and maintain all types of prompts including standalone prompts and meta-prompts for AI-to-AI workflows.

Generate images using OpenAI's GPT Image model. Use for creating images from text prompts. Keywords: image, generate, openai, gpt-image, dalle, picture, art, create.

Use to validate process documentation. Apply TDD to skill writing: RED (run without skill, document failures) → GREEN (write skill) → REFACTOR (close loopholes). Test under pressure: time constraints, sunk cost, exhaustion, authority.

Result backend configuration patterns for Celery including Redis, Database, and RPC backends with serialization, expiration policies, and performance optimization. Use when configuring result storage, troubleshooting result persistence, implementing custom serializers, migrating between backends, optimizing result expiration, or when user mentions result backends, task results, Redis backend, PostgreSQL results, result serialization, or backend migration.

Enhanced guide for creating effective skills with automatic packaging workflow. 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 tool integrations. Automatically offers to package and provide download links for immediate installation.

Analyze existing PHPStan error logs without running analysis. Use when:- User says "fix the phpstan errors" (after manually running PHPStan)- User says "what phpstan errors do I have?"- User points to specific log file- PHPStan was run outside Claude's contextDelegates to fixer agent (sonnet) to find logs, parse errors, and implement fixes.Does NOT execute PHPStan - use phpstan-runner for that.

Apply the Explore → Plan → Code → Commit workflow for tasks like implementing features, fixing bugs, refactors, or adding integrations. Use this when the user wants changes in a repo and wants higher reliability (read first, plan then commit/PR). Emphasize subagents for investigation during Explore, and use "think / think hard / think harder / ultrathink" during Plan when alternatives exist.

Cross-language testing patterns: unit, integration, E2E testing strategies.Use when: writing tests, setting up test infrastructure, or improving coverage.Triggers: "test", "testing", "unit test", "integration test", "e2e","coverage", "mock", "fixture".

Recursive meta-skill that detects workflow patterns and automatically generates Claude Code agent definitions. Use when user invokes /create-agent or patterns warrant new agent.

Process and generate multimedia content using Google Gemini API. Capabilities include analyze audio files (transcription with timestamps, summarization, speech understanding, music/sound analysis up to 9.5 hours), understand images (captioning, object detection, OCR, visual Q&A, segmentation), process videos (scene detection, Q&A, temporal analysis, YouTube URLs, up to 6 hours), extract from documents (PDF tables, forms, charts, diagrams, multi-page), generate images (text-to-image, editing, composition, refinement). Use when working with audio/video files, analyzing images or screenshots, processing PDF documents, extracting structured data from media, creating images from text prompts, or implementing multimodal AI features. Supports multiple models (Gemini 2.5/2.0) with context windows up to 2M tokens. | Sử dụng khi: AI, LLM, vision, embedding, phân tích hình ảnh, Gemini API.