Skill UI

The `hf-mem` skill estimates the necessary VRAM or system memory required to load Safetensors or GGUF model weights for inference directly from the Hugging Face Hub. It uses efficient HTTP Range requests, avoiding local downloads. Users can determine if a model fits on their GPU or instance, including optional estimations for the KV cache and setting limits like maximum sequence length.

Automates the creation and publication of polished, functional Gradio demos on Hugging Face Spaces for any user-provided LoRA model. This skill handles the complex workflow of designing the optimal UI, selecting the correct inference pipeline (e.g., text-to-image, video-to-video), and generating all necessary files (app.py, requirements.txt, README.md) to deploy a complete, ready-to-use AI playground.

A comprehensive utility for building reusable, command-line scripts and utilities to interact with the Hugging Face API. It enables complex data workflows by allowing users to chain, pipe, and process data fetched from models, datasets, and various endpoints. Ideal for automation, enrichment, and advanced data pipelines.

This skill is designed to accurately convert complex, Hugo-managed documentation sites and content into standard Markdown format. It goes beyond generic knowledge by inspecting the entire local Hugo repository structure, including `hugo.toml`, custom shortcodes, layouts, and content files. It preserves the original content's semantic meaning while materializing Hugo-specific constructs into clean Markdown.

This skill manages the end-to-end lifecycle of cloud test execution using HyperExecute for platforms like LambdaTest/TestMu AI. It provides comprehensive functionality to analyze projects, generate and validate complex YAML configurations, run CLI jobs, and debug failures. Ideal for integrating robust, scalable, and matrix-based testing into CI/CD pipelines.

This skill guides the structured development process for implementing new features or fixes based on Product Requirement Documents (PRDs) or predefined issues. It emphasizes best practices, including Test-Driven Development (TDD), regular type-checking, running comprehensive test suites, and completing the cycle with a formal review process.

This skill performs a deep comparison of two code versions to verify semantic equivalence, going beyond simple syntax diffing. It uses semi-formal tracing techniques to trace both versions against shared specifications and boundary conditions (e.g., null, zero, max/min). Ideal for confirming if a refactor, rewrite, or migration has unintentionally altered the code's behavior, answering the question: "Did I break anything?"

This skill generates a detailed, step-by-step, interprocedural execution trace for a given piece of code and specific input scenario. It resolves name ambiguities, tracks type transitions, and highlights non-obvious behaviors (like implicit coercions or side effects). Use it when a user is confused about how code behaves, allowing for a deep dive into the actual execution path to clarify assumptions versus reality.

This skill provides an autonomous, repository-wide pipeline for auditing and fixing complex logical inconsistencies. It systematically executes stages including health checks, deep review, locating divergences, generating fixes, and performing diff-verification until the codebase is deemed clean. Ideal for maintaining high code quality and resolving deep, systemic bugs across entire projects.

This skill is designed to identify subtle and complex logic bugs within a single code file or function. It employs a semi-formal execution tracing methodology, systematically breaking down the suspected bug into five stages: Premises (expected state), Trace (execution path), Divergence (actual deviation), Trigger (reproduction case), and Remedy (fix). It is ideal for scenarios where users suspect code is wrong but cannot specify a concrete failure point.

Access comprehensive real-time and historical market data across major global exchanges, including HK, US, A-share, and Singapore. This skill provides quotes, K-line charts, order book depth, trade ticks, market sentiment indices, and sophisticated cross-market analysis tools like ADR premium and FX carry trade frameworks. Ideal for quantitative analysis, technical trading, and monitoring global financial movements.

This library provides a comprehensive framework for managing bounded AI workflows (loops). Users can analyze existing codebases or work histories to discover opportunities for repeating, automated agent processes. It guides the user through finding published templates, auditing and repairing weak agents, adapting parameters, or crafting entirely new, robust automation designs, ensuring clear triggers, verification, and mandatory stopping conditions.