SuperAPI AI Directory

writing-plugin-readme

Writing & research

View repository

Updated 10-16 04:12

About this skill

Writing Plugin README

Overview

A guide for writing clear, readable, and highly professional README documents for plugins, tools, and sub-resources within this repository.

When to Use

  • You need to write or update the main repository README.md for this plugin.
  • You are documenting a new script, tool, or reference catalog.
  • You want to ensure the README has consistent structure, copy-pasteable commands, and clear setup guidelines.

Essential README Checklist

A professional README for a tool or plugin must include these sections:

  1. Header/Title: Clear title and a 1-2 sentence high-level summary of what the plugin/tool does.
  2. Features: Bulleted list of key features or capabilities.
  3. Prerequisites & Installation: Explicit dependencies (e.g., Python packages, yt-dlp, SQLite3) and command-line instructions to set up.
  4. Usage Guide: Real-world CLI examples with expected input/output.
  5. Directory / Architecture Map: Visual chart or list mapping key files and directories.
  6. Contributing & License: Basic contribution rules and license terms.

Core Templates

1. Main Plugin README Template

# Plugin Name

A brief 1-2 sentence description explaining the purpose of this plugin and who it is built for.

## Features
- **Feature A**: Description of what Feature A does.
- **Feature B**: Description of what Feature B does.

## Prerequisites & Installation
Ensure you have the following installed on your system:
- Python 3.10+
- [yt-dlp](https://github.com/yt-dlp/yt-dlp) (for media downloads)

Install skills to your agent via the Skills CLI:
```bash
# Install all skills globally
npx skills add zenithth/anibon-stream-synthesis --all -g

# Or install a specific skill (e.g. anibon-timestamper)
npx skills add zenithth/anibon-stream-synthesis --skill anibon-timestamper -g -y

Quick Start & Usage

Provide a simple, copy-pasteable command to get started immediately:

python3 scripts/run_tool.py --argument value

Options

  • --verbose: Enable detailed logs.
  • --limit: Max number of items to process.

File Structure

  • scripts/: Python utility scripts.
  • skills/: Nested agent sub-skills.
  • tests/: Automated unit and integration tests.

License

Distributed under the MIT License. See LICENSE for details.


## Formatting Rules
- **Use Code Blocks**: Always wrap commands, options, and code snippets in fenced code blocks with language tags (`bash`, `python`, `json`).
- **Use Alert Boxes Strategically**: Use GitHub alerts (`> [!NOTE]`, `> [!IMPORTANT]`, `> [!WARNING]`) to highlight environment requirements or licensing constraints.
- **Checklist Formatting**: Use standard markdown checkboxes (`- [ ]`) for installation tasklists or configurations.

## Common Mistakes
- **Vague Commands**: Documenting a command like `python run.py` without specifying the relative path or required arguments. Always write exact paths (e.g., `python3 scripts/run_tool.py`).
- **Missing Prerequisites**: Forgetting to mention external CLI tools (like `ffmpeg` or `yt-dlp`) that are not standard Python libraries.
- **Placeholder Text**: Leaving incomplete sections or placeholder bracket text like `<insert details here>`. Fill in or omit.