Skill UI

Statsmodels is a robust Python library providing comprehensive tools for advanced statistical analysis. It supports a wide range of models, including Ordinary Least Squares (OLS), Generalized Linear Models (GLM like Poisson and Logistic), Weighted Least Squares (WLS), and complex time series forecasting (ARIMA). Use this skill for rigorous econometrics, detailed diagnostic testing, and generating publication-ready statistical inference.

This skill utilizes the TimesFM foundation model for zero-shot forecasting of any univariate time series. It predicts future values with calibrated probabilistic intervals without requiring custom model training. It includes a mandatory preflight system checker to ensure adequate RAM, GPU, and disk resources for safe, local inference across various data types (sales, sensor, energy).

This skill utilizes the Kronos model to perform advanced financial time-series forecasting. It generates quantitative base predictions and then integrates news sentiment analysis to adjust and refine the forecast, making it an essential tool for finance professionals needing actionable market insights.

This comprehensive guide details advanced strategies for managing time-series data and data retention in vector databases (like Qdrant). It is designed for scenarios where only recent data is relevant (e.g., social media feeds, news articles). Three core methods are presented: Shard Rotation (recommended for natural time boundaries), Collection Rotation (for per-period configuration needs), and Filter-and-Delete (for simple cleanup). The guide also covers hot/cold tiering to optimize performance.

This workflow automates the creation of step-by-step documentation guides using the TechSmith Snagit COM API and PowerShell. It first captures a series of sequential screenshots for a guide, and then uses a batch process to open and annotate these images within the Snagit Editor, ensuring consistency and efficiency in creating professional tutorials.

This technique is designed for the final verification stage when analyzing a candidate TV series. Instead of accepting initial matches, users must perform targeted searches combining the show name and specific character details (e.g., birth year, specific plot event, relationship). This ensures that the candidate show satisfies all complex constraints, eliminating false positives and confirming the accuracy of the plot details.

This technique is used to locate a specific TV series by leveraging the known filmography of its creator (director, writer, or showrunner). When the question provides clues that the creator also worked on a distinct, identifiable project (e.g., a trilogy or an award-winning miniseries), use that secondary work as a powerful search anchor. Identify the person first using their other works, then trace back through their complete credits to pinpoint the target series.

Generate a cohesive three-panel social media carousel (1080x1080), designed for cinematic storytelling. Perfect for creating multi-part, on-brand content series for platforms like Instagram, LinkedIn, or X. The skill ensures the panels read as one continuous narrative while maintaining individual visual appeal.

This extension tracks how often your brand is cited or mentioned across leading Large Language Models (LLMs) like ChatGPT and Perplexity. It provides detailed time-series data, including citation rates and trend deltas (week-over-week, month-over-month). It is essential for SEO professionals to monitor brand authority and visibility within the rapidly evolving landscape of generative AI search results.

A comprehensive 10-page pitch deck template designed for fundraising and presenting to venture capitalists (VCs). It covers all essential sections, including problem identification, solution overview, market size (TAM/SAM/SOM), traction data visualization, business model, go-to-market strategy, and the funding ask. Ideal for startups seeking seed or Series A funding.

A comprehensive skill designed to fetch raw market data (Open, High, Low, Close, Volume) for a given symbol. It normalizes the time series data into relative OHLCV vectors, converts these vectors into 64-dimension embeddings, and stores them in a memory store. Crucially, it indexes the resulting vectors using the HNSW algorithm, enabling fast and precise nearest-neighbor search for pattern detection and similarity analysis in financial time series.

This skill solves the mean-variance portfolio optimization problem (Σ · x = μ) using the highly efficient Conjugate Gradient (CG) method. It significantly outperforms legacy Neumann series methods, providing state-of-the-art speedup (40-60x) and accuracy for calculating optimal investment weights. It is designed for quantitative finance and risk management applications where computational performance is critical.