Skill UI

This comprehensive script suite provides methods to optimize GPU cloud computing expenses on Vast.ai. It guides users through selecting the most cost-efficient GPU for their workload, leveraging interruptible (spot) instances, and implementing critical lifecycle management safeguards. Features include detecting idle compute resources, auto-destroy timers, and calculating total cost burn rates, ensuring AI model training and inference jobs stay within budget.

A comprehensive guide for deploying Machine Learning training jobs and inference services on the Vast.ai GPU cloud. This skill covers the entire deployment lifecycle, including optimizing Docker images for performance, creating automated provisioning scripts using the Vast.ai CLI, implementing robust data transfer strategies (SCP, rsync), and performing post-deployment health checks.

ZAI CLI provides comprehensive access to Z.AI's advanced capabilities, including image/video analysis, real-time web search, web page extraction, and GitHub code exploration. It is designed for users needing multi-modal content analysis, error diagnosis, and complex data processing through a unified command-line interface. Ideal for developers and analysts requiring robust, API-driven AI functionality.

A comprehensive directory designed for discovering, researching, and comparing autonomous AI agents, frameworks, and tools across diverse ecosystems. This skill helps users map the current AI agent landscape, identify existing solutions, gather best practices, and pinpoint market gaps before initiating custom development or workflow design.

Translates high-level creative intent into detailed cinematography directives for Veo3, Kling, and Luma video models, orchestrating framing, movement, lighting, and lens choices before generating cinematic scenes.

DiffDock uses diffusion-based deep learning to predict protein-ligand binding poses, confidence, and batch virtual screening workflows for structure-based drug design.

Hypogenic automates LLM-driven hypothesis generation and testing over tabular datasets, blending literature insights with empirical analysis to accelerate discovery in domains like deception detection or AI content evaluation.

Hardware-agnostic quantum ML framework enabling gradient-based training of quantum circuits, hybrid quantum-classical models, and seamless integration with PyTorch/JAX/TensorFlow alongside device portability across IBM/Google/Rigetti/IonQ.

PufferLib is a high-performance library designed for advanced reinforcement learning simulations and training. It excels in parallel environment simulation, offering optimized vectorization and native support for multi-agent systems. Utilizing algorithms like PuffeRL (optimized PPO+LSTM), it achieves millions of steps per second, significantly accelerating research and deployment across single-agent and multi-agent environments.

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.

Hierarchical Agent Memory (HAM) scopes Claude Code agent context by generating per-directory CLAUDE.md files, tracking token savings, and surfacing analytics via a dashboard so multi-folder projects avoid repeated full scans.

Transform briefs into cinematic videos via Seedance 2.0, guiding text/image/video/audio inputs through director-level directives, multimodal tagging, and synchronized sound for 4-15s outputs.