Skill UI

A powerful Python skill for adding professional callouts, arrows, and labels to various images, including screenshots, diagrams, and architecture documents. It allows users to visually highlight specific areas or changes. The advanced `annotate.py` module provides algorithmic label placement, optimizing for readability and preventing overlaps. Ideal for creating clear PR descriptions, documenting changes, or building annotated demo assets.

A comprehensive guide on embedding before-and-after screenshots directly into pull request descriptions. This skill helps reviewers visualize UI changes, layout adjustments, and data visualizations instantly, without needing to check out the branch. Includes best practices for image sizing, structured Markdown patterns, and specific API/upload guidance for Azure DevOps and GitHub.

This comprehensive skill provides a full workflow for creating high-quality, annotated animated GIFs and screen recordings. It supports capturing multi-step UI interactions using Playwright, assembling dynamic GIFs with imageio, applying variable frame timing, and adding sophisticated annotations (including fade-in effects) for clear documentation, pull request descriptions, or release note demos.

A comprehensive guide for securing container images and runtime environments. Covers best practices from Dockerfile authoring (minimal base images, multi-stage builds, non-root users) through advanced runtime enforcement (seccomp, AppArmor) and Kubernetes security controls. Essential for production hardening and maintaining supply chain integrity.

LiteParse is a fast, open-source tool for local, layout-aware text extraction from PDFs, Office files, and images. It provides structured JSON output with precise bounding boxes and supports OCR, making it ideal for building robust RAG systems, citation grounding, and multimodal agents. All processing is done locally without needing cloud APIs.

Generate professional and modern presentations using the 2slides AI API. This tool supports creating decks from raw text, matching a specific design style via a reference image, summarizing documents, adding AI voice narration, and exporting results as PDFs or images. Use it for any 'make slides' or 'create a deck' request.

The RunAPI CLI serves as the dedicated execution layer for interacting with RunAPI's suite of advanced AI models. It allows developers to programmatically generate multimodal content, including images, videos, and music/audio, directly from the terminal or CI/CD pipeline. Users can submit one-off jobs, manage asynchronous tasks, and pass complex JSON request bodies to various services like Suno, Claude, and Gemini, making it essential for robust AI automation workflows.

Generates a highly structured, machine-parseable shooting script for short videos based on a specified topic, style, and duration. It outputs a detailed shot list including image prompts, video prompts, voiceovers, and on-screen text, making it ideal for automated AI video generation workflows.

A comprehensive guide on pre-caching assets, fonts, images, and initial data in Flutter and Flutter Web. This technique significantly improves the perceived performance by avoiding loading flashes, font swaps, and jank, ensuring a smooth user experience, especially on startup. Best practices emphasize caching only the content the user will see in the immediate next screens.

A powerful skill for generating high-quality single images from text prompts (Text-to-Image) or modifying existing images (Image-to-Image). It utilizes OpenRouter, supporting various aspect ratios and resolutions, making it ideal for concept art, product rendering, or any single visual asset creation. The tool handles sophisticated parameters like model choice and failure fallbacks.

Generates high-quality, short video clips (3-15s) using the Seedance 2.0 model. This tool accepts structured English prompts, an optional first-frame image, and an optional identity/style reference image for continuity. It supports dual backends (OpenRouter, Volcengine, BytePlus) to maximize accessibility and reliability for cinematic shot generation.

Designed to capture a fully functional, standalone static HTML file from any modern web application (React, Next.js, Vue, etc.). It resolves all computed styles and inlines CSS and images (as Base64) into the output, ensuring the captured view can be shared or archived without external dependencies. Supports both automated snapshotting via Puppeteer and interactive capture.