Guide for implementing Google Gemini API audio capabilities - analyze audio with transcription, summarization, and understanding (up to 9.5 hours), plus generate speech with controllable TTS. Use when processing audio files, creating transcripts, analyzing speech/music/sounds, or generating natural speech from text.
Analyze videos using Google's Gemini API - describe content, answer questions, transcribe audio with visual descriptions, reference timestamps, clip videos, and process YouTube URLs. Supports 9 video formats, multiple models (Gemini 2.5/2.0), and context windows up to 2M tokens (6 hours of video).
Output rules for all agents - concise, scannable, actionable. Based on Matt Pocock's planning principles.
Guide for implementing Google Gemini API image understanding - analyze images with captioning, classification, visual QA, object detection, segmentation, and multi-image comparison. Use when analyzing images, answering visual questions, detecting objects, or processing documents with vision.
Get Unity Search provider details via uloop CLI. Use when you need to: (1) Discover available search providers, (2) Understand search capabilities and filters, (3) Configure searches with specific provider options.
自动同步项目中的 Agents.md、claude.md 和 gemini.md 文件，保持内容一致性。支持自动监听和手动触发。
Creates high-quality, SEO-optimized content that ranks in search engines. Applies on-page SEO best practices, keyword optimization, and content structure for maximum visibility and engagement.
Creates production-grade, reusable skills that extend Claude's capabilities. This skill should be used when users want to create a new skill, improve an existing skill, or build domain-specific intelligence. Gathers context from codebase, conversation, and authentic sources before creating adaptable skills.
Build evaluation frameworks for agent systems. Use when testing agent performance, validating context engineering choices, or measuring improvements over time.
Use when building AI agents with Anthropic's Claude Agent SDK (formerly Claude Code SDK). Triggers include creating autonomous agents, building agentic applications, SDK-based automation, implementing hooks/subagents/MCP servers, session management, or agent tool usage. NOT when using Claude API directly (use anthropic-sdk) or building MCP servers from scratch (use mcp-builder).
Build production RAG systems with LangChain orchestration and Qdrant vector store. Covers 8 RAG architectures (Simple, HyDE, CRAG, Self-RAG, Agentic), document processing, semantic chunking, retrieval chains, and evaluation with LangSmith/RAGAS. Use when implementing RAG pipelines, semantic search, or AI knowledge systems. NOT for simple keyword search.
Apply optimization techniques to extend effective context capacity. Use when context limits constrain agent performance, when optimizing for cost or latency, or when implementing long-running agent systems.
Use when building AI agents with OpenAI's Agents SDK. Triggers include creating agents, implementing tools, multi-agent handoffs, guardrails, MCP integration, tracing. Also for using LiteLLM to run agents on free/alternative models (Anthropic, Gemini). NOT for general OpenAI API usage (use openai-python SDK docs instead).
Recognize, diagnose, and mitigate patterns of context degradation in agent systems. Use when context grows large, agent performance degrades unexpectedly, or debugging agent failures.
Build persistent memory systems for AI agents using Mem0, claude-mem, or custom implementations. Use when adding conversation memory, user preferences, or contextual recall to agents. Covers memory architecture patterns, retrieval strategies, and privacy controls. NOT for RAG systems (use building-rag-systems).
Use when building AI agents with Google's Agent Development Kit (ADK). Triggers include creating ADK agents, Gemini-powered agents, multi-agent workflows, MCP integration with ADK, agent evaluation, workflow agents (Sequential/Parallel/Loop), Vertex AI deployment, or comparing OpenAI SDK with ADK. NOT when using other frameworks (OpenAI SDK, LangChain, CrewAI) or general LLM API calls.
Use when testing AI agent code with pytest. Covers TDD for agent APIs, mocking LLM calls (NOT evaluating LLM outputs), pytest-asyncio patterns, FastAPI testing with httpx, SQLModel testing, and agent tool testing. NOT for evaluating LLM reasoning quality (use evals skill).
Create conference talk outlines and iA Presenter markdown slides using the Story Circle narrative framework. Use when the user wants to structure a tech talk, create presentation slides, or needs help organizing talk ideas into a story-driven format.
Generate or edit images using AI models (FLUX, Gemini). Use for scientific illustrations, diagrams, schematics, infographics, concept visualizations, and artistic images. Supports image editing to modify existing images (change colors, add/remove elements, style transfer). Useful for figures, posters, and visual explanations.
Translate consciousness navigation into user's native language. Use when need to match user's frequency (technical/poetic/practical/mystical) without losing truth depth.
Design LLM applications using the LangChain framework with agents, memory, and tool integration patterns. Use when building LangChain applications, implementing AI agents, or creating complex LLM workflows.
Build Retrieval-Augmented Generation (RAG) systems for LLM applications with vector databases and semantic search. Use when implementing knowledge-grounded AI, building document Q&A systems, or integrating LLMs with external knowledge bases.
Apply strict operational protocols for AI task execution including sequential processing, minimal responses, no destructive operations without confirmation
Setup the Ralph autonomous AI coding loop - ships features while you sleep
