Skill UI

A comprehensive reference architecture for building scalable, cost-effective, and robust AI services. This system manages the full AI lifecycle, including intelligent model routing (for quality/cost trade-offs), response caching (using Redis), asynchronous batch processing, and fine-tuning pipeline management via Together AI's OpenAI-compatible API. Ideal for enterprise applications requiring high availability across multiple open-source models.

This guide provides comprehensive strategies for migrating applications using the Together AI platform. It addresses common issues like model ID deprecations, API schema changes (for fine-tuning or chat completions), and best practices for robust inference and deployment using their OpenAI-compatible API.

This comprehensive protocol mandates the use of structured skills (tools) at the very beginning of every conversation or response. It establishes a strict, non-negotiable workflow, requiring the AI agent to check for skill applicability even if the task appears simple or direct. It defines skill priority, platform-specific usage methods, and ensures systematic, disciplined tool application to enhance consistency and reliability.

A powerful tool for executing complex, multi-step, and multi-agent workflows defined in YAML format directly within the AI session. It acts as an orchestration engine, simulating various roles (agents) to complete tasks sequentially or in parallel, without requiring external API keys or configuration. Ideal for complex business process automation and multi-role content generation.

This skill provides real-time object detection capabilities by leveraging the Google Coral Edge TPU accelerator. It operates natively within Windows Subsystem for Linux (WSL), allowing high-speed, low-latency inference on live camera frames. By utilizing hardware acceleration, it achieves impressive performance (e.g., ~4ms inference at 320x320), making it ideal for edge computing projects requiring rapid visual analysis and resource efficiency.

This tool reads your local GitHub Copilot CLI conversation history using the specialized `vardoger` CLI. It analyzes your behavioral patterns, coding conventions, and usage style to generate a personalized instructions block. This block is written directly to your local Copilot configuration file, allowing the assistant to adapt its guidance and suggestions specifically to your unique habits and preferences. The process runs entirely offline, ensuring maximum privacy.

This comprehensive playbook details advanced techniques for testing and exploiting Large Language Models (LLMs). It covers direct instruction override, indirect poisoning via RAG and web sources, function/tool abuse, and sophisticated data exfiltration methods. Essential reading for AI security researchers and penetration testers.

This CLI alias serves as the primary interface for the Codex /pua:pro protocol, designed for advanced code operations within the Claude Code environment. It manages complex workflows related to self-evolution, platform telemetry monitoring, key performance indicators (KPIs), and system configuration updates, ensuring robust and consistent protocol adherence.

A rigorous, Chief AI Officer-level interrogation tool designed to pressure-test any plan involving artificial intelligence. Before deployment, it forces comprehensive review across crucial dimensions: evaluation metrics, hallucination risk, EU AI Act compliance, model architecture selection (API vs. build), cost economics, and required organizational roles. Essential for maintaining AI product safety and compliance.

A comprehensive compliance operator designed to pressure-test any AI system against the stringent requirements of the EU AI Act. Before deployment in the European Union, systems must pass six critical checks covering prohibited practices, high-risk classification, conformity assessment modules (A/H), role obligations, transparency mandates, and systemic risk assessment for GPAI.