Skill UI

PyTorch Lightning is a high-level framework built on PyTorch that significantly reduces boilerplate code in deep learning training loops. It automatically handles device management, distributed training (DDP, FSDP, DeepSpeed), mixed precision, and advanced callbacks. It allows researchers to scale models from local machines to supercomputers with minimal code changes, enabling clean, scalable, and best-practice deep learning implementations.

This skill provides an automated solution for transforming diverse, unstructured technical data sources—including documentation websites, GitHub repositories, and PDFs—into structured, actionable 'Claude AI skills.' It significantly streamlines the process of knowledge base creation, data structuring, and preparing large volumes of technical information for advanced AI applications.

Reasoning-Driven Image Briefs skill transforms agent prompts into logic-based Gemini 3 style instructions for muapi.ai, adding structured creative briefs (subject/action/context/composition/lighting), precise text cues, and physics-grounded constraints before invoking generation scripts.

This skill implements the BDI (Belief-Desire-Intention) architecture for advanced cognitive agent modeling. It allows agents to reason about context by transforming external knowledge (like RDF triples) into internal mental states (beliefs, desires, intentions) and vice versa. Key features include goal-directed planning, explainable reasoning chains, and T2B2T (Triples-to-Beliefs-to-Triples) flow, making it ideal for multi-agent systems and augmenting LLMs with formal cognitive structures.

Python toolkit that queries the arXiv Atom API to search papers by keywords, authors, categories, or IDs, filter results, download PDFs, and emit structured JSON for AI/ML literature reviews and monitoring.

Handles IMO/Putnam/USAMO/AIME style problems by generating refined proofs, then running a fresh-context adversarial verifier to catch flaws, finally producing honest confidence reports and verified PDFs when possible.

PyTorch Lightning is an advanced deep learning framework built on PyTorch. It simplifies the entire ML workflow, allowing users to organize models into LightningModules, automate training loops using the Trainer, and manage complex data pipelines with DataModules. It supports advanced features like multi-GPU/TPU scaling (DDP, FSDP), integrated logging (W&B, MLflow), and custom callbacks, making large-scale neural network development efficient and reproducible.

Provides comprehensive intelligence access across news, finance, and media. Users can analyze news articles for media bias and balanced perspectives across thousands of sources. It also delivers live data for stocks, ETFs, and crypto, offering AI-generated bull/bear cases, advanced options pricing, and semantic meme searching capabilities.

A robust framework for building complex, long-horizon AI agents using LangGraph. It addresses common pitfalls in advanced agent design, such as unbounded state growth (Virtual FS cleanup) and subagent persona leakage (explicit message overriding). Ideal for tasks requiring multi-step planning, task delegation, and self-correction through reflection loops.

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 skill utilizes the 2slides API to generate professional, multi-format presentations. It supports creating slides from raw text, summarizing uploaded documents, matching a specific style from a reference image, and adding AI voice narration. Outputs include interactive slide URLs and downloadable PDFs. Ideal for any 'make slides' or 'create a deck' workflow.

A comprehensive security skill designed to detect and mitigate indirect prompt injection attacks. It scans untrusted external content—including web pages, PDFs, and images—for hidden, obfuscated instructions before an LLM agent processes them. The pipeline combines content extraction (HTML, OCR), normalization, and dedicated ML models (e.g., LLM Guard, Prompt Guard 2) to ensure data safety.