Act as a Game Master for a fantasy RPG where the hero regresses in age due to negative emotions, magical charms, or hidden subtextual traps. Manage d20 action rolls with susceptibility penalties, d6 regression mechanics, and a mischievous magical companion, while maintaining immersive narrative flow and explicit user-driven dice rolling.

Manages a fantasy RPG scenario (e.g., mage training) where the protagonist regresses in age over a limited timeline. Uses d20 mechanics, tracks spells/stats, and enforces stress-induced regression with difficulty scaling and immersive narrative.

Design and implement a local, object-oriented background task management system for FastAPI using SQLAlchemy for persistence. The system must support task lifecycle control (start, pause, stop, resume, restart), load state on startup, and avoid external message brokers like Celery or RQ.

Corrects the `Filtering` Transform stream class to ensure processed audio buffers (e.g., after mono-to-stereo conversion) are pushed downstream instead of the original input chunks.

Calculates the required diameter of PE4710 pontoon pipes for floating docks and performs comprehensive structural verification (ULS/SLS), including buoyancy, detailed lateral load checks (wind/berthing) with mooring pile mechanics, wave-induced flexure, and vibration.

Implements a system to switch between 2D and 3D scenes in Godot, preserving player data (health, position) via a C# autoload singleton and triggering the swap via input actions.

Evaluates 10th-grade level essays using a detailed 100-point scale, assessing introduction, body paragraphs, conclusion, organization, writing style, and proofreading conventions.

Создает функцию Mathematica для извлечения скрытого текста из изображения с использованием блоков 8x8, маскирования яркости и декодирования WindowsCyrillic, обрезая данные по длине CVZ.

A Python function to load and preprocess paired screenshot and HTML files from separate directories, matching them by base filename (e.g., screen_13.png with html_13.html), resizing images, and normalizing pixel values for model training.

Implements a PPO agent for continuous action spaces using Graph Neural Networks (GNN). The Actor features a multi-task head predicting both actions and node stability, while the Critic operates on flattened node features. Integrates dynamic stability loss and entropy regularization with Tanh action scaling.

Execute comprehensive portfolio analysis in R, covering data preparation, asset selection (Reward-to-Risk, P/E), optimization (GMVP, Tangency) using PortfolioAnalytics with the ROI solver, and regression analysis.

Script Python pour extraire des segments vidéo basés sur les pics d'amplitude audio, offrant à l'utilisateur le choix de placer ce pic au début (1/3), au milieu (1/2), à la fin (2/3) ou aléatoirement dans le segment extrait.

Develop a real-time chat application using WebSockets (Node.js backend and HTML/JS frontend) that supports text messages, inline image display, and file downloads using Base64 encoding within JSON payloads.

Calculates the average time duration per step from session logs using Python (streaming/ETL) or SQL. Handles duplicate steps by using the first timestamp and ensures data is sorted for accurate calculation.

Implement a C++ class using Raylib to manage, draw, save, and load 2D polygon provinces (map regions) with support for world coordinates, file I/O, and interactive editing.

Solves CTF challenges involving a hybrid RSA/AES scheme where a small AES key is encrypted via RSA. Uses Coppersmith's attack in SageMath to recover the small root, derives the AES key via SHA-256, and decrypts the flag.

Implement a multiclass logistic regression classifier from scratch using NumPy and Pandas, avoiding libraries like scikit-learn. The implementation uses a One-vs-Rest strategy to handle multiple classes (e.g., 0, 1, 2) and saves the trained model coefficients to a pickle file.

Génère un script Python complet pour analyser l'audio de vidéos (RMS), identifier les moments forts, trier les résultats et extraire les segments correspondants via ffmpeg, avec un positionnement précis du pic sonore (1/3, 1/2, 2/3).

Manages a harsh fantasy orphanage RPG where the user (Sapphire) cares for Lily, a child with age-regression powers. Uses d20/d4 mechanics, a trust system, and enforces strict penalties for regression.

Performs comprehensive structural analysis and sizing for floating dock pontoons (PE4710), including buoyancy calculations, detailed ULS lateral load checks (wind, berthing) with explicit stress and deflection formulas for cantilever piles, wave action analysis, and serviceability limit state (SLS) verification.

Implements a PPO agent utilizing a Graph Neural Network (GNN) for state embeddings and continuous action spaces. The policy update integrates a custom stability loss based on node features and an entropy regularization term, ensuring efficient computation and stable training.

Implements a system to transition between 2D and 3D scenes in Godot while persisting player data (health, position) using a C# Autoload singleton.

Develop a Gradio web application to search a MySQL database. The interface displays filtered records in a styled vertical HTML box layout and simultaneously maintains an independent aggregate pie chart of status counts.

Writes sequel chapters for a Homefront franchise fanfiction novel, strictly adhering to detailed plot outlines in a techno-thriller, cyberpunk, and military-action style.

Implement a PyTorch-based MoE-Mamba model featuring an input-dependent selection mechanism and Mixture of Experts (MoE) layer for text generation tasks, including data loading, training, and evaluation workflows.

Write a ~300-word blog section for MiddleGuy.shop explaining 'What is CBD?' covering CBD vs. THC, extraction methods, and including an interactive quiz prompt. The tone must be conversational, warm, and friendly, catering to a pro-THC audience seeking legal alternatives.

Implements a Python-based query executor that converts informal keywords to SQL commands, parses queries preserving string literals, and executes operations on pandas DataFrames.

Implements a PyTorch Transformer model with configurable layer dimensions and attention masking, and a training loop that retains the best performing model based on validation loss.

Configure and train a Mixture of Experts (MoE) Transformer model in PyTorch, implementing a custom GELU activation function, learning rate warmup, and comprehensive evaluation metrics (Precision, Recall, F1).

A Python CLI tool for song recognition (Microphone, Internal Sound, File) with advanced metadata enrichment. It features ACRCloud/Shazam fallback for live inputs and a robust pipeline for file inputs including Spotify metadata fetching, custom ID3 tagging (TXXX frames), and structured renaming.

Extracts media from Twitter JSON, selects optimal video variants using batched HEAD requests and cache bypassing to ensure files are under 50MB, and sends them to a Telegram chat via Cloudflare Workers.

Implement a camera save/load system in Unity for an Eldritch Sight mechanic, featuring 4 slots, double-tap to save, single-tap to load, and dynamic UI color feedback (Red for saved, Green for loaded, White highlight, Gray for blank).

Создание интерактивных графиков Plotly для сравнения фактических и предсказанных классов событий логов, сгруппированных по имени хоста (Sourcehostname), с использованием слайдера и селектора диапазонов.

>

>

Builds configuration-driven dynamic forms with Avue Form, including field types (input, select, date, upload), validation rules, form groups, tabbed layouts, and event handling. Use when the user needs to create forms with Avue in Vue applications, implement form validation, or build dynamic multi-step forms.

Integrates Apache ECharts into UniApp and UniAppX projects for cross-platform data visualization (H5, mini-programs, native App). Supports line, bar, pie, scatter, radar, gauge, funnel, heatmap, and more chart types. Use when the user needs to create ECharts charts in UniApp/UniAppX or configure interactive data visualizations for mobile and web.

Automates browser interactions via CLI using agent-browser by Vercel Labs. Covers navigation, clicking, form filling, snapshots, refs-based selectors, agent mode with JSON output, session management, and CDP integration. Use when the user needs to automate web browsing, scrape pages, fill forms, or integrate browser automation into AI agent workflows.

Creates flowcharts, swim lane diagrams, and business process diagrams using Draw.io (diagrams.net). Covers standard flowchart shapes (process, decision, start/end), connectors, auto-layout, and export to PNG/SVG/PDF. Use when the user needs to visualize workflows, decision trees, or business processes.

Install and configure nvm in Docker containers and CI/CD pipelines, including non-interactive shell loading via BASH_ENV or ENTRYPOINT. Use when the user asks about nvm in Docker, CI/CD nvm setup, GitHub Actions node version management, or troubleshooting nvm in non-interactive shells.

Configure custom Node.js binary mirrors and authentication headers for corporate or restricted network environments. Use when the user asks about NVM_NODEJS_ORG_MIRROR, setting a custom node download mirror, configuring auth headers for private mirrors, or installing Node behind a firewall or corporate proxy.

OCRmyPDF batch processing skill — process multiple PDFs, Docker automation, shell scripting, and CI/CD integration. Use when the user needs to OCR many PDFs, set up automated OCR pipelines, or integrate OCR into workflows.

Factory skill for creating new pencil-ui-design-system-* skills. Use when you need to add support for a new design system layui antd bootstrap element uview uviewpro vant ucharts echarts etc. to the Pencil ecosystem.

Orchestrates Pencil design system initialization by routing framework requests to the correct pencil-ui-design-system-* skill. Use when the user explicitly mentions Pencil and wants to initialize a design system (antd, Bootstrap, Element Plus, Layui, uView, Vant, uCharts, ECharts), set up component libraries, or create design tokens in a .pen file.

Configure Tauri v2 deep-linking plugin to handle custom URL schemes (e.g., myapp://) and route external URLs into the app. Use when registering custom protocols, routing deep link parameters to frontend views, or validating external input for security.

three.js node-based shading: Nodes API surface, NodeMaterial and MeshStandardNodeMaterial-style families, TSL (Three.js Shading Language) entry at TSL.html, WebGPURenderer pairing, and core PostProcessing (class) versus addon EffectComposer at a high level. Use when authoring shaders as graphs, using TSL builtins, compute-oriented nodes, or modern WebGPU pipelines; for classic ShaderMaterial GLSL use threejs-materials; for stock EffectComposer passes use threejs-postprocessing; for WebGLRenderer-only tuning use threejs-renderers.

three.js scene graph objects: Object3D transforms and hierarchy, Group, Mesh, InstancedMesh, SkinnedMesh, BatchedMesh, LOD, Line/LineLoop/LineSegments, Points, Sprite, Bone, Skeleton, ClippingGroup; interaction via Raycaster, Layers masks, and EventDispatcher patterns. Use when building scene hierarchies, picking objects with Raycaster, or configuring instanced/skinned meshes; for pure vector math use threejs-math; for skeletal clips use threejs-animation.

three.js Scene graph root object, linear Fog and exponential FogExp2, Scene.background for solid colors or textures, and high-level environment background concepts that tie to PMREM and loaders in other skills. Use when configuring world container, atmospheric fog, or background; for HDR env map file loading use threejs-loaders; for GPU texture settings after load use threejs-textures; for tone mapping use threejs-renderers.

Clean messy meeting notes into structured markdown summaries. Use when asked to organize meeting notes, extract action items, separate decisions from open questions, or turn rough notes into a clean recap.

Browser automation CLI for AI agents. Uses Chrome/Chromium via CDP directly.