answerzhao/agent-skills
14 skills2.7k installs364 starsGitHub
Install
npx skills add https://github.com/answerzhao/agent-skillsSkills in this repo
1Web ReaderThe web-reader skill. Implement web page content extraction capabilities using the z-ai-web-dev-sdk. Use this skill when the user needs to scrape web pages, extract article content, retrieve page metadata, or build applications that process web content. Supports automatic content extraction with title, HTML, and publication time retrieval. **Reference Scripts**: Example test scripts are available in the directory for quick testing and reference. The API automatically handles content extraction, providing clean, structured data from any web URL. **IMPORTANT**: z-ai-web-dev-sdk MUST be used in backend code only. Import it as shown in the examples below. This is ideal for quick content scraping, testing URLs, or simple automation tasks. Extract main article content and metadata 3. The workflow follows the source SKILL.md contract with progressive reference loading, clear trigger phrases, and practical steps developers can apply directly in agent sessions.1.7kinstalls2Vlmvlm is a Claude Code agent skill from answerzhao/agent-skills that guides backend integration of vision-language models through the z-ai-web-dev-sdk package and an optional npx z-ai vision CLI. The skill documents createVision chat completions with image_url and base64 payloads, stream and JSON output flags, and a reference scripts/vlm.ts example for testing multimodal flows. Developers reach for vlm when building agents that must analyze screenshots, describe uploaded images, or route visual inputs alongside text in conversational tools. The skill enforces backend-only SDK usage and covers three supported media content types for vision chat. It pairs naturally with broader agent-tooling work where terminal scripts and typed API handlers must stay outside the client bundle.334installs3AsrAgent skill for automatic speech recognition: transcribing audio to text, supporting streaming partial results, preprocessing input, and feeding transcripts into LLM-driven agent workflows.193installs4TtsTTS is a skill that implements text-to-speech using the z-ai-web-dev-sdk audio API. It documents backend code and a z-ai CLI to convert text into spoken audio, with support for multiple voices, adjustable speed from 0.5 to 2.0, and wav, pcm, and mp3 formats, plus constraints like a 1024-character input limit and a chunking helper for long text. Developers use it to add voice output or generate audio files.104installs5Web SearchWeb-search is a skill that implements web search using the z-ai-web-dev-sdk web_search function. It documents backend code and a z-ai CLI to query the web and retrieve current information, returning structured results with URL, title, snippet, host name, rank, date, and favicon, and supporting result-count and recency-day filters. Developers use it to add real-time web search or current-information retrieval to an application.94installs6Image GenerationImage-generation is a skill that implements AI text-to-image generation using the z-ai-web-dev-sdk. It shows backend code and a z-ai CLI for creating images from text prompts, supports seven fixed sizes across square, portrait, and landscape, and returns base64-encoded images. Developers use it to add image creation to an application or generate assets like hero images and logos from the command line.41installs7Frontend DesignFrontend-design is a skill that turns vague UI style requirements into production-ready frontend code using a design-token system. It enforces a tokens-first workflow covering semantic colors, typography, spacing, radius, and shadow scales, targets stacks like React with Tailwind and shadcn/ui, and bundles CSS and TypeScript examples plus Tailwind and globals.css templates. Developers use it when building websites, web apps, dashboards, or landing pages.36installs8LlmLLM is a skill for building conversational AI features using the z-ai-web-dev-sdk chat completions API. It documents backend patterns for system prompts, multi-turn conversation history, chain-of-thought thinking, and a ConversationManager class, plus a z-ai CLI for quick one-off completions. Developers use it to add chatbots, AI assistants, or text generation to an application.35installs9PdfPdf is a skill for programmatically processing PDF documents. It covers extracting text and tables, creating new PDFs, merging and splitting documents, OCR of scanned pages, watermarking, and filling PDF forms, using Python libraries like pypdf, pdfplumber, and reportlab plus command-line tools. It bundles form-handling scripts and a reference file. Developers use it to generate or analyze PDFs at scale or to fill in forms.23installs10XlsxXlsx is a skill for creating, editing, and analyzing spreadsheets. It documents workflows using pandas for data analysis and openpyxl for formulas and formatting, requires Excel formulas rather than hardcoded values and zero formula errors, and covers financial-model conventions like color coding and number formatting, using LibreOffice to recalculate formulas. Developers use it to build or modify .xlsx, .csv, and related files while preserving formulas.21installs11DocxDocx is a skill for programmatically creating and manipulating Microsoft Word .docx documents. It documents the JavaScript/TypeScript docx library for building documents with text formatting, tables, images, headers and footers, hyperlinks, and a table of contents, and it bundles OOXML schemas with pack, unpack, and validate scripts for working at the raw XML level. Developers use it to generate Word documents at scale.20installs12Video GenerationVideo-generation is a skill that implements AI video generation using the z-ai-web-dev-sdk. It documents a two-step asynchronous model that creates a task then polls for the result, supporting text-to-video, image-to-video, first-last-frame transitions, optional AI audio, and custom resolution, frame rate, and duration, plus a z-ai CLI that auto-polls. Developers use it to produce video content programmatically.20installs13Canvas DesignCanvas-design is a skill for creating static visual art such as posters and design pieces, output as .png and .pdf files. It works in two steps: first it writes a design philosophy or aesthetic movement as a markdown manifesto, then it expresses that philosophy visually on a canvas. It ships a library of open-license fonts and instructs the model to produce original designs rather than copy existing artists.19installs14PptxPptx is a skill for programmatically building PowerPoint presentations. It documents converting HTML slides into .pptx with accurate positioning using an html2pptx library and PptxGenJS, with rules for slide dimensions, supported HTML elements, and web-safe fonts, and it bundles OOXML schemas with pack, unpack, and validate scripts. Developers use it to generate presentations from HTML at scale.18installs