Skill UI

This tool reads a CSV containing three columns (x, y_baseline, y_ours). It uses matplotlib to generate a two-line plot visualizing the results and saves the final output as a PDF file. It is specifically designed for visualizing performance comparisons in scientific papers.

Use PaperJSX to automate the creation of professional documents, including PPTX presentations, DOCX reports, XLSX spreadsheets, and PDF files. By providing a structured JSON specification, this tool generates ready-to-use files, eliminating manual formatting and ensuring consistency across large-scale reporting and content generation pipelines.

A robust toolkit for deterministic, structural PDF manipulation. This skill allows users to perform programmatic tasks such as extracting text and tables from reports, merging multiple PDF files, splitting documents by page ranges, filling out form fields, or generating entirely new PDFs from structured data (e.g., JSON). It is designed for explicit, byte-level operations, distinguishing it from LLM-based natural language rewriting tools.

This utility parses structured short-drama scripts (AI-video-script format) and generates a time-coded SRT subtitle file. It reads cumulative shot durations and voiceover texts to create highly accurate, timestamped subtitle cues. It operates purely on text processing, making it ideal for automated video production pipelines, especially between script merging and final subtitle burning.

This skill automates the migration of existing frontend code (including React, Vue, Angular, and Next.js components) into a structured Stitch Design. It orchestrates the extraction of a self-contained HTML file, analyzes the source code to build a comprehensive design system document, and finally uploads both assets into your Stitch project for iteration and improvement.

Analyzes any frontend codebase (React, Vue, Svelte, Angular, plain CSS, Tailwind) to document the complete design system. It extracts crucial visual elements—including color palettes, typography, spacing scales, and component patterns—directly from source files, without needing to build, render, or run the application. Ideal for design audits, visual language documentation, or migrating a project's identity into a design tool.

This skill captures a fully self-contained static HTML snapshot from any running web application (e.g., React, Vue, Next.js). Using Puppeteer, it renders the live DOM, automatically inlining all CSS styles and converting images, assets, and backgrounds to base64 data URIs. This ensures the generated HTML file is completely self-sufficient and requires no external network connection, making it perfect for sharing UI states, creating documentation, or preparing assets for offline viewing.

This skill acts as a mobile engineering bridge, converting web designs provided by Stitch into high-quality, production-ready React Native components. It handles complex tasks including translating HTML/CSS layouts into React Native primitives, extracting design tokens (colors, spacing) into a central theme file, and ensuring adherence to native component architectural rules. Ideal for developers integrating web-based prototypes into mobile applications.

This skill enables the upload of various local assets, including images (PNG, JPG, WEBP), extracted HTML, and design markdown files, directly to a designated Stitch project. Crucially, it bypasses the token limits associated with standard MCP tool calls by using a dedicated HTTP upload script. Use this when visual assets or large design documentation need to be integrated into your Stitch design system.

This utility uses ffmpeg's libass filter to permanently burn an SRT subtitle file into an MP4 video stream. It handles video re-encoding (H.264) while copying the audio track directly, ensuring high quality. Supports customization of font name, size, margin, and video encoding parameters (CRF, preset), making it essential for the final subtitling step in video content pipelines.

This utility reads the raw content of a UTF-8 text file and emits it verbatim to stdout. It is specifically designed for meta-skills that require round-tripping an artifact through disk, allowing users to manually edit structured files (like scripts or YAML) during review pauses. Unlike standard built-in readers that add line numbering, this method preserves the original bytes, making it suitable for downstream parsers.

This powerful utility concatenates multiple numbered MP4 segments (e.g., 1_*.mp4, 2_*.mp4) into a single, cohesive video file. It supports optional fade transitions and normalizes video parameters, including resolution, FPS, and codec, ensuring a professional and unified output. It is ideal for stitching together short clips produced by automated workflows into a final reel or presentation.