Skill UI

Designs and builds production-ready real-time voice agents with OpenAI Realtime, Vapi, Deepgram, ElevenLabs, LiveKit, and WebRTC, optimizing for latency, audio quality, and responsive UX.

Stable Baselines3 is a robust, PyTorch-based library providing production-ready implementations of core reinforcement learning algorithms (PPO, SAC, DQN, etc.). It facilitates the training of single-agent RL models, supports the creation of custom Gymnasium environments, and handles advanced parallel training via vectorized environments. It is ideal for rapid prototyping and rigorous RL experiments.

This comprehensive guide provides step-by-step instructions for integrating the Google Gemini API into applications using Python and Node.js. It covers the entire lifecycle, from basic text generation to advanced use cases like multimodal inputs (text + image), real-time streaming responses, function calling/tool use, and multi-turn chat. Includes best practices for model selection, error handling, and production deployment.

A comprehensive guide to building production-grade evaluation systems for Large Language Models (LLMs). This skill covers advanced methodologies, including Direct Scoring and Pairwise Comparison, alongside critical techniques for mitigating systematic biases (e.g., Position Bias, Length Bias). Learn how to select appropriate metrics and structure prompts to ensure reliable, consistent, and objective quality assessment across various AI applications.

This comprehensive guide outlines best practices for building reliable and robust AI agent harnesses. It covers optimizing action spaces, defining structured tools, managing observation formats, and implementing essential error recovery contracts. Use these principles to dramatically improve agent completion rates, manage context budgets effectively, and transition AI prototypes into stable, production-grade autonomous systems.

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.

Comprehensive guide to LLM Operations (LLMOps). Covers advanced topics like Retrieval-Augmented Generation (RAG), vector database management (Pinecone, Chroma), embeddings, fine-tuning, prompt engineering, and cost optimization. Essential for bridging the gap between AI prototypes and reliable, scalable, production-grade AI systems.

Creates photorealistic or stylized images via Stability AI (SD3.5/Ultra/Core) with text-to-image, img2img, inpainting, upscale, background removal, erase, and search-replace modes for artistic or production-ready output.

PydanticAI is a robust Python framework that integrates Pydantic's type-checking and validation capabilities into Large Language Models (LLMs). It allows developers to build production-ready AI agents that guarantee structured, validated outputs, moving beyond simple raw strings. Key features include supported tool calling, dependency injection for testability, and multi-model compatibility across major providers like OpenAI, Anthropic, and Gemini. Use this when reliability and correct data structure are paramount.

An autonomous factory for generating production-ready, cross-platform AI agent skills. It accepts raw, unstructured human input—such as transcripts, PDFs, screenshots, vague ideas, or code snippets—and reverse-engineers the complete workflow logic. It bridges the gap between human intent and functional agent specifications, allowing users to automate complex processes without writing code.

Helps teams treat prompts like production code by providing registry, eval, and governance workflows for versioning, testing, and promoting prompts safely across environments.

A robust, production-tested pattern for coordinating multiple specialized AI agents. The orchestrator decomposes complex tasks, routes them to the most suitable specialist, prevents redundant or conflicting work (anti-duplication), and enforces mandatory quality gates (e.g., running tests, checking file diffs) to verify results before marking a task as complete. It also includes heartbeat monitoring for task progress.