Skill UI

This comprehensive guide covers all aspects of integrating the Google Gemini API into your applications using Python or Node.js. It details basic text generation, advanced multimodal inputs (text, image, audio), streaming responses, function calling/tool use, multi-turn chat implementation, and production best practices like model selection and robust error handling.

A framework for building production-ready, low-latency, multimodal AI agents. It allows developers to orchestrate complex pipelines that seamlessly handle real-time audio, video, and text data. Key capabilities include integrating multiple AI services (LLMs, STT, TTS), managing conversation context, and enabling tool use and function calling for advanced, interactive AI applications.

This skill guides the development of EarLLM One, a complex Android application. The app captures real-time audio from Bluetooth earbuds, transcribes the voice using STT, sends the text input to a Large Language Model (LLM) via a robust pipeline, and plays the response back using TTS. It requires deep knowledge of Kotlin, Android development, Bluetooth profiles (A2DP/HFP), and advanced audio routing.

This blueprint guides the creation of F.R.I.D.A.Y., a tactical, low-latency voice assistant inspired by Iron Man. It utilizes the specialized Pipecat framework to orchestrate a comprehensive, real-time audio pipeline: STT (OpenAI Whisper/GPT-4o) -> LLM (Gemini) -> TTS (OpenAI Nova). Ideal for developers needing to build advanced, multi-provider, conversational voice applications.

A comprehensive skill for ingesting, analyzing, and manipulating multi-modal video and audio content. Supports real-time desktop capture, live stream monitoring (RTSP), and file uploads (URL/local). Capabilities include advanced indexing (visual, semantic, temporal), precise moment searching, professional timeline editing (subtitles, overlays, dubbing), transcoding, and generating actionable alerts. Ideal for content creators, monitoring systems, and complex media workflows.

A comprehensive skill for generating multimodal media content using fal.ai models. It supports text-to-image (Nano Banana), text/image-to-video (Veo, Kling, Seedance), and text/video-to-audio (CSM-1B, ThinkSound). Use this tool for any creative project requiring high-quality generated images, videos, or sound effects.

A comprehensive toolset for digital forensics, designed to detect, extract, and analyze covert data hidden within images, audio, and video files. It utilizes advanced steganalysis techniques (LSB, DCT, frequency domain analysis) and specialized tools (StegDetect, zsteg, binwalk) to uncover hidden communications channels, making it crucial for investigations involving data exfiltration or insider threats.

This skill detects AI-generated deepfake audio used in voice phishing (vishing) attacks. It functions by extracting critical spectral features (including MFCCs, spectral centroid, and zero-crossing rate) and passing them through advanced machine learning models (like Random Forest and Gradient Boosting). The system supports batch forensic analysis, provides confidence scores, and generates detailed reports to verify the authenticity of recorded voice calls, making it essential for cybersecurity incident response and fraud investigation.

Transformers.js allows developers to run state-of-the-art machine learning models directly within JavaScript/TypeScript environments (browser, Node.js). It supports a comprehensive range of tasks including NLP (text generation, summarization, translation), Computer Vision (image classification, object detection), and Audio processing. This capability enables building fully client-side, multimodal AI applications without requiring a dedicated Python backend or server infrastructure.

A comprehensive guide covering best practices for creating videos using Remotion and React. Learn how to incorporate advanced features such as 3D rendering (Three.js), complex animations, audio synchronization, data charting, subtitles, and sophisticated scene transitions. This skill covers 29 domain-specific rules for professional video pipelines.

A CLI integration for Lark Minutes, enabling users to search, download media, upload audio/video for transcription, update titles, replace speakers, and manage AI tasks.

This skill transforms written content, such as articles and reports, into engaging, two-host conversational podcast audio. It first structures the content into a detailed JSON script (dialogue format) and then uses advanced Text-to-Speech synthesis to generate the final MP3 file. Supports both English and Chinese languages, making content accessible for audio-only listening.