DOCSEARCHbyAlgolia
Submit Docs

open-agent-hub

repository·main·Indexed 21 days ago

https://github.com/guanyang/open-agent-hub

A lightweight CLI tool (oah) for managing and activating modular capabilities—Skills, Agents, and Slash Commands—for AI coding assistants including Claude Code, Cursor, Trae, Gemini CLI, and others. It provides unified capability management via local symlinks, upstream synchronization, and a compatibility matrix for project-level and system-level activation.

Tokens
951.4K
Snippets
1.9K
Records
3.3K
Agent score
72%

What's inside open-agent-hub

  1. Overview of PPTX creation, editing, and analysis

    main

    The pptx skill handles all .pptx and .potx file operations. Depending on your goal, choose one of these three primary approaches:

    • Create a new deck: Write a pptxgenjs script (Node.js).
    • Edit an existing deck or build from a template: Unzip the archive, edit the XML files (e.g., ppt/slides/slideN.xml), and re-zip.
    • Read content: Use markitdown to extract text (one block per slide under <!-- Slide number: N --> markers) or python scripts/thumbnail.py to generate a visual grid of slides.

  2. Overview of Antigravity Skills categories

    main

    The Antigravity Skills suite (v2.1.0) contains 50 total skills categorized into several functional domains. Available domains include:

    • Algorithmic Art: Creative and generative art capabilities.
    • Brand Guidelines: Maintaining brand consistency.
    • Canvas Design: Visual layout and design tasks.
    • Doc Co-authoring: Collaborative document creation.
    • DOCX (Word Processing): Handling Microsoft Word files.
    • Frontend Design: UI/UX and frontend development tasks.
    • Internal Comms: Communication-related automation.
    • NotebookLM (Knowledge Base Q&A): Querying knowledge bases.
    • MCP Builder (MCP Server Construction): Building Model Context Protocol servers.
    • PDF Processing: Extracting and manipulating PDF data.

  3. Use the remotion-multimedia skill for audio and video processing

    main

    The remotion-multimedia skill (version 4.0.504) provides capabilities for interacting with Mediabunny, a multimedia library designed for handling audio and video files within a browser environment.

    This skill allows you to perform the following tasks:

    • Get audio duration: Retrieve the length of an audio file in seconds.
    • Get video dimensions: Retrieve the width and height of a video file.
    • Get video duration: Retrieve the length of a video file in seconds.

    For a detailed overview of Mediabunny's full capabilities, refer to its technical specification at: https://mediabunny.dev/llms.txt.

  4. Use the baoyu-xhs-images skill for social media infographics

    main

    The baoyu-xhs-images skill generates infographic image card series (1-10 cards) optimized for social media platforms like Xiaohongshu (小红书), WeChat (微信), or Weibo. It breaks down complex content into visually engaging, cartoon-style cards.

    Triggering the skill: You can trigger this skill by mentioning terms like:

    • 小红书图片, 小红书种草, 小绿书
    • 微信图文, 微信贴图
    • image cards, 图片卡片
    • baoyu-xhs-images or requesting social media infographic series.

  5. Use the baoyu-format-markdown skill

    main

    The baoyu-format-markdown skill transforms plain text or markdown into well-structured, reader-friendly markdown by adding frontmatter, titles, summaries, headings, bold text, lists, and code blocks.

    Core Principle: The skill only adjusts formatting and fixes obvious typos. It never adds, deletes, or rewrites original content.

    Triggering the skill: Use this skill when you need to:

    • "format markdown"
    • "beautify article"
    • "add formatting"
    • Improve article layout

    Output: The formatted content is saved to {filename}-formatted.md.

  6. Use the webapp-testing skill for local web application testing

    main

    The webapp-testing skill provides a toolkit for interacting with and testing local web applications using Playwright. It allows you to verify frontend functionality, debug UI behavior, capture screenshots, and view browser logs.

    Core Workflow

    1. Identify the target: If it is a static HTML file, read the file directly to find selectors. If it is a dynamic web application, you must manage the server lifecycle.
    2. Manage the server: Use the scripts/with_server.py helper to ensure your application servers are running before your tests execute.
    3. Write automation: Create Python scripts using Playwright to navigate, inspect, and interact with the application.

    Best Practices

    • Use sync_playwright() for synchronous automation scripts.
    • Always close the browser at the end of your script.
    • Wait for networkidle: For dynamic applications, always use page.wait_for_load_state('networkidle') before attempting to inspect the DOM or interact with elements to ensure JavaScript has finished executing.
    • Use descriptive selectors: Prefer text=, role=, CSS selectors, or IDs.

  7. Use the mediabunny skill for multimedia handling

    main

    The mediabunny skill provides capabilities for handling audio and video files within a browser environment using the Mediabunny library. It is primarily used for extracting metadata from multimedia files.

    Key capabilities include:

    • Getting audio duration (in seconds)
    • Getting video dimensions (width and height)
    • Getting video duration (in seconds)

  8. Convert X (Twitter) content to Markdown

    main

    The baoyu-danger-x-to-markdown skill converts X (Twitter) tweets, threads, and articles into Markdown files with YAML front matter. It supports full content extraction for X Articles.

    Supported URLs:

    • https://x.com/<user>/status/<id>
    • https://twitter.com/<user>/status/<id>
    • https://x.com/i/article/<id>

    Output Format: Markdown files include a YAML front matter block containing the url, author, tweetCount, and coverImage.

    File Structure: By default, files are saved to ./x-to-markdown/{username}/{tweet-id}/{content-slug}.md.

    # Example usage with a URL
${BUN_X} {baseDir}/scripts/main.ts https://x.com/user/status/123

  9. Use the Remotion Best Practices router

    main
    The remotion-best-practices skill acts as a central router for all Remotion-related tasks within the Open Agent Hub. Instead of searching for specific Remotion instructions, use this skill to identify which specialized sub-skill to load based on your specific goal (e.g., creating a project, working with maps, or deploying a SaaS).

  10. Understand the `context-fundamentals` skill

    main

    The context-fundamentals skill is a conceptual foundation for context engineering. It is used to explain and reason about the core mechanics of how language models process information, such as the anatomy of a context window, attention mechanics, and the importance of context quality over quantity.

    When to use this skill:

    • For conceptual explanations and onboarding.
    • For reasoning about design decisions from first principles.
    • For writing or reviewing documentation that needs to ground guidance in underlying mechanics.

    When NOT to use this skill (use specialized skills instead):

    • To diagnose attention failures or 'lost-in-the-middle' issues: use context-degradation.
    • To implement token-efficiency tactics (masking, partitioning, budgets): use context-optimization.
    • To compress long sessions into summaries: use context-compression.
    • To manage large tool outputs or scratchpads via files: use filesystem-context.
    • To decide the architecture of an LLM project: use project-development.

  11. Use remotion-multimedia for audio and video processing

    main

    The remotion-multimedia skill (version 4.0.504) provides capabilities for interacting with Mediabunny, a multimedia library designed for handling audio and video files within a browser environment.

    Key capabilities include:

    • Retrieving audio duration in seconds.
    • Retrieving video dimensions (width and height).
    • Retrieving video duration in seconds.

  12. Use the XLSX skill for spreadsheet tasks

    main

    The xlsx skill is designed for any task where a spreadsheet file is the primary input or output.

    Use this skill to:

    • Create or edit .xlsx, .xlsm, .xltx, .csv, or .tsv files.
    • Perform data operations: adding columns, computing formulas, formatting, charting, or cleaning messy tabular data.
    • Convert between different tabular file formats.
    • Read/Analyze existing spreadsheets.

    Do NOT use this skill if the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration.