Intellectronica Agent Skills

repository·main·Indexed 43 days ago

https://github.com/intellectronica/agent-skills

A collection of specialized capabilities designed to extend AI agent functionality. Includes skills for interacting with external APIs and services such as AnkiConnect, Google Workspace (gog-cli), Context7, GitHub Copilot SDK, and OpenAI's GPT Image 1.5, as well as utilities for Mermaid diagram rendering, markdown conversion via markitdown, semantic code search with mgrep, and git repository initialization.

Tokens
127.7K
Snippets
405
Records
568
Agent score
74%

What's inside Agent Skills

  1. Overview of the here-be-git skill

    main

    The here-be-git skill initializes a git repository in the current directory. It can optionally include agent-specific commit instructions and a .gitignore file.

    The skill is triggered when a user provides intents such as:

    • "here be git"
    • "init git"
    • "initialise git"
    • Or any other indication of wanting to set up version control in the current directory.
  2. Overview of the notion-api skill capabilities

    main

    The notion-api skill provides comprehensive instructions for interacting with the Notion API via REST calls. It is designed to handle requests involving:

    • Content Management: Reading, creating, updating, or deleting pages, databases, blocks, and comments.
    • Core API Logic: Authentication, endpoint management, pagination, and error handling.
    • Best Practices: Implementation of Notion-specific best practices for reliable API interaction.
  3. Overview of the nano-banana-pro skill

    main

    The nano-banana-pro skill allows agents to generate and edit images using Google's Nano Banana Pro (Gemini 3 Pro Image) API.

    Use cases:

    • Text-to-image generation: When a user asks to "generate", "create", or "make" an image.
    • Image-to-image editing: When a user asks to "modify", "change", "alter", or "update" an image, or references an existing image file for modification (e.g., "change the background", "replace X with Y").

    Key Capabilities:

    • Supports both generation and editing.
    • Configurable resolution: Defaults to 1K, but supports 2K and 4K for high-resolution requirements.

    Important Usage Note: Do NOT attempt to read the image file manually before using this skill. Instead, pass the image directly using the --input-image parameter.

  4. Overview of the gog-cli skill

    main

    The gog-cli skill enables command-line operations for Google Workspace. Use this skill when you need to automate or interact with Google services via an agent.

    Supported Google Workspace services include:

    • Gmail (search/send)
    • Calendar (events)
    • Drive (files)
    • Docs, Sheets, Slides, Forms
    • Apps Script
    • Chat
    • Classroom
    • Contacts, Tasks, People, Groups
    • Admin (Workspace administration)
    • Keep
    • Authentication and configuration

    When to use this skill: Trigger this skill when the user mentions gog, gogcli, Google CLI, or specific tasks like Gmail search, Calendar events, Drive files, or Workspace admin automation.

  5. Overview of copilot-sdk capabilities

    main

    The copilot-sdk skill provides support for GitHub Copilot SDK workflows across multiple languages, including Node.js/TypeScript, Python, Go, .NET, and Java. It enables capabilities such as:

    • Setup and Authentication
    • Permissions management
    • Streaming events
    • Custom tools and custom agents
    • MCP (Model Context Protocol) servers
    • Hooks and skills
    • Session persistence
  6. Overview of mgrep-code-search skill

    main

    The mgrep-code-search skill provides semantic code search using mgrep for efficient codebase exploration. It is designed for natural language semantic search, making it ideal for finding features, understanding intent, and exploring unfamiliar code.

    When to use: Use this skill when searching or exploring codebases that contain more than 30 non-gitignored files or involve complex nested directory structures. It serves as a semantic complement to traditional grep or ripgrep workflows.

  7. Overview of the Tavily skill

    main
    The Tavily skill provides capabilities for web search, extraction, mapping, crawling, and research via Tavily's REST API. It is particularly useful when an agent needs web search capabilities that are not provided by built-in tools, or when the LLM-friendly format provided by Tavily is preferred for research tasks.
  8. Overview of the markdown-converter skill

    main

    The markdown-converter skill uses markitdown to convert various document and file formats into Markdown. This is useful for preparing data for LLM processing or text analysis. Supported formats include:

    • Documents: PDF, Word (.docx), PowerPoint (.pptx), Excel (.xlsx, .xls), EPubs
    • Web/Data: HTML, CSV, JSON, XML, YouTube URLs
    • Media: Images (with EXIF/OCR), Audio (with transcription)
    • Archives: ZIP archives