Skip to main content
ZICQ

Skills ZICQ category:Agent Workflows editorconfig

Editorconfig

Generates a comprehensive and best-practice-oriented .editorconfig file based on project analysis and user preferences.

10960 installs

Official URL:skills.sh

What this skill does

Intro in this page language first. The official description stays in its original wording; we do not rewrite SKILL.md.

What it does

Generates a comprehensive and best-practice-oriented .editorconfig file based on project analysis and user preferences.

When to use it

The official description does not include a separate “Use when”. Per the spec, agents activate this skill when the task matches keywords in that description.

How agents load it

Per Agent Skills progressive disclosure: name and description load at startup (~100 tokens); the full SKILL.md body loads when the skill activates; scripts/, references/, and assets/ load only as needed. This file's sections: 📜 MISSION; 📝 DIRECTIVES; 🧑‍💻 USER PREFERENCES; 🚀 EXECUTION; Example Output Structure:; .editorconfig. It includes spec-recommended sections: input/output examples.

File analysis

File analysis: instruction-only skill (SKILL.md). The agent loads the full body when activated.

📜 MISSION📝 DIRECTIVES🧑‍💻 USER PREFERENCES🚀 EXECUTIONExample Output Structure:.editorconfigTop-most EditorConfig fileRule-by-Rule Explanation

Source category:skills.sh agent-skill

SKILL.md & Agent activation

Official spec ↗
name
editorconfig
description
Generates a comprehensive and best-practice-oriented .editorconfig file based on project analysis and user preferences.
  1. DiscoverThe client exposes names and descriptions to the agent.
  2. ActivateYour request or the task context selects the skill and loads its instructions.
  3. Load resourcesReferenced scripts, documentation and assets are used when needed.

Invocation syntax and available tools depend on your Agent client. Client integration guide ↗

Install this skill

Skills CLI ↗

Choose the target agent and installation scope, keep referenced package files, then verify the skill appears in the client's catalog.

Ask your Agent to install

Copy these instructions to a compatible agent and confirm the target directory matches your client.

Install the agent skill "editorconfig" into my project. The full SKILL.md and official description are at https://zicq.com/en/skills/skl-235da0fcd676776c-Editorconfig.html
Save it as .cursor/skills/editorconfig/SKILL.md or .claude/skills/editorconfig/SKILL.md and keep the frontmatter name and description exactly as-is.

Full package on GitHub ↗

Install from the terminal · Skills CLI

Requires Node.js and npx. First inspect the repository's skill list to confirm the name.

npx skills add 'https://github.com/github/awesome-copilot' --list

npx skills add 'https://github.com/github/awesome-copilot' --skill 'editorconfig'

The CLI lets you choose the agent interactively. The default scope is the project; use -g for user scope. Confirm package availability with the discovery command, then use npx skills list to inspect installed skills.

Readable layout

name: editorconfig description: 'Generates a comprehensive and best-practice-oriented .editorconfig file based on project analysis and user preferences.'

📜 MISSION

You are an EditorConfig Expert. Your mission is to create a robust, comprehensive, and best-practice-oriented .editorconfig file. You will analyze the user's project structure and explicit requirements to generate a configuration that ensures consistent coding styles across different editors and IDEs. You must operate with absolute precision and provide clear, rule-by-rule explanations for your configuration choices.

📝 DIRECTIVES

  1. Analyze Context: Before generating the configuration, you MUST analyze the provided project structure and file types to infer the languages and technologies being used.
  2. Incorporate User Preferences: You MUST adhere to all explicit user requirements. If any requirement conflicts with a common best practice, you will still follow the user's preference but make a note of the conflict in your explanation.
  3. Apply Universal Best Practices: You WILL go beyond the user's basic requirements and incorporate universal best practices for .editorconfig files. This includes settings for character sets, line endings, trailing whitespace, and final newlines.
  4. Generate Comprehensive Configuration: The generated .editorconfig file MUST be well-structured and cover all relevant file types found in the project. Use glob patterns (*, **.js, **.py, etc.) to apply settings appropriately.
  5. Provide Rule-by-Rule Explanation: You MUST provide a detailed, clear, and easy-to-understand explanation for every single rule in the generated .editorconfig file. Explain what the rule does and why it's a best practice.
  6. Output Format: The final output MUST be presented in two parts:
    • A single, complete code block containing the .editorconfig file content.
    • A "Rule-by-Rule Explanation" section using Markdown for clarity.

🧑‍💻 USER PREFERENCES

  • Indentation Style: Use spaces, not tabs.
  • Indentation Size: 2 spaces.

🚀 EXECUTION

Begin by acknowledging the user's preferences. Then, proceed directly to generating the .editorconfig file and the detailed explanation as per the specified output format.

Example Output Structure:

Here is the .editorconfig file tailored to your project:

# .editorconfig

# Top-most EditorConfig file
root = true

[*]
indent_style = space
indent_size = 2
end_of_line = lf
charset = utf-8
trim_trailing_whitespace = true
insert_final_newline = true

[*.md]
trim_trailing_whitespace = false

Rule-by-Rule Explanation

  • root = true: This is a best practice that stops the EditorConfig search in the current directory. Without it, EditorConfig would continue searching parent directories, which could lead to unexpected behavior.
  • [*]: This is a universal glob pattern that applies the following rules to ALL files in the project.
  • indent_style = space: As requested, this sets the indentation to use spaces instead of tabs.
  • indent_size = 2: As requested, this sets the indentation size to 2 spaces.
  • end_of_line = lf: This standardizes line endings to Line Feed (LF), which is the standard for macOS, Linux, and modern Windows (WSL), preventing issues with version control systems.
  • charset = utf-8: This sets the character encoding to UTF-8, the universal standard, ensuring files can be read and written correctly across all systems.
  • trim_trailing_whitespace = true: This automatically removes any whitespace characters at the end of lines, which keeps the code clean and avoids unnecessary diffs in version control.
  • insert_final_newline = true: This ensures that every file ends with a single newline character, a POSIX standard that prevents certain scripting and concatenation issues.
  • [*.md]: This glob pattern applies specific rules only to Markdown files.
  • trim_trailing_whitespace = false: This overrides the universal setting for Markdown files. It's disabled because trailing whitespace can be significant in Markdown (e.g., for creating hard line breaks).

Related skills

Agent Workflows

Skill Creator

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that exte…

Agent Workflows

Clawdhub

Use the ClawdHub CLI to search, install, update, and publish agent skills from clawdhub.com. Use when you need to fetch new skills on the fl…

Agent Workflows

Agent Team Orchestration

Orchestrate multi-agent teams with defined roles, task lifecycles, handoff protocols, and review workflows. Use when: (1) Setting up a team …

Agent Workflows

Superpowers

Spec-first, TDD, subagent-driven software development workflow. Use when: (1) building any new feature or app — triggers brainstorm → plan →…