AgentSkillsCN

readme-generator

通过分析代码库结构，自动生成软件项目的完整 README.md 文件。

SKILL.md
--- frontmatter
name: readme-generator
description: Generates comprehensive README.md files for software projects by analyzing codebase structure
allowed-tools: ["Read", "Glob", "Grep", "Write"]
version: 1.0.0
author: GLINCKER Team
license: Apache-2.0
keywords: [documentation, readme, markdown, project]

README Generator

Automatically generates professional, comprehensive README.md files by analyzing your project structure, dependencies, and code patterns.

What This Skill Does

This skill helps you create high-quality README files by:

  • •Analyzing project structure and identifying key components
  • •Detecting programming languages and frameworks
  • •Finding configuration files (package.json, requirements.txt, etc.)
  • •Identifying test frameworks and CI/CD setup
  • •Generating appropriate sections with relevant content
  • •Following README best practices

Instructions

When generating a README, follow these steps:

1. Project Discovery

First, analyze the project structure:

  • •Use Glob to find key files: package.json, pyproject.toml, Cargo.toml, go.mod, etc.
  • •Use Glob to identify main source directories
  • •Use Read to examine configuration files
  • •Use Grep to find test files and CI configuration

2. Content Analysis

Based on findings, determine:

  • •Project type (library, application, CLI tool, etc.)
  • •Primary programming language(s)
  • •Dependencies and frameworks
  • •Build and test commands
  • •License type (from LICENSE file)

3. README Generation

Create a README.md with these sections (adapt based on project type):

Required Sections:

  • •Title and Description: Project name and one-line summary
  • •Features: Key functionality (if applicable)
  • •Installation: How to install/set up
  • •Usage: Basic usage examples
  • •License: License information

Optional Sections (include if relevant):

  • •Prerequisites: Required software/tools
  • •Development: How to set up for development
  • •Testing: How to run tests
  • •Contributing: Contribution guidelines
  • •API Documentation: For libraries
  • •Screenshots: For applications with UI
  • •Roadmap: Future plans
  • •Acknowledgments: Credits and thanks

4. Writing Style

Use this style for generated READMEs:

  • •Clear, concise language
  • •Active voice
  • •Code blocks with proper syntax highlighting
  • •Badge shields for status indicators (if CI/CD detected)
  • •Emoji sparingly (only if user requests)
  • •Professional but approachable tone

5. Output

Present the generated README to the user and offer to:

  • •Write it to README.md
  • •Make adjustments based on feedback
  • •Add additional sections

Examples

Example 1: Python Project

User Request: "Generate a README for this Python project"

Workflow:

  1. •Glob for Python files: **/*.py
  2. •Read pyproject.toml or setup.py
  3. •Check for requirements.txt, Pipfile
  4. •Look for test files in tests/ or *_test.py
  5. •Generate README with:
    • •Installation via pip
    • •Python version requirements
    • •Virtual environment setup
    • •Testing with pytest/unittest

Example 2: Node.js Project

User Request: "Create a README for my npm package"

Workflow:

  1. •Read package.json for name, description, scripts
  2. •Identify framework (React, Vue, Express, etc.)
  3. •Check for TypeScript (tsconfig.json)
  4. •Look for test configuration (Jest, Mocha)
  5. •Generate README with:
    • •npm/yarn installation
    • •Available scripts
    • •API documentation (for packages)
    • •Usage examples

Configuration

This skill adapts to project type:

Project TypeKey FilesFocus Areas
Pythonpyproject.toml, setup.pypip install, virtual env
Node.jspackage.jsonnpm install, scripts
RustCargo.tomlcargo build, features
Gogo.modgo get, modules
GenericNoneBasic structure

Tool Requirements

  • •Read: Examine configuration and source files
  • •Glob: Find relevant files across project
  • •Grep: Search for patterns (tests, CI, etc.)
  • •Write: Create the README.md file

Limitations

  • •Cannot include screenshots (user must add manually)
  • •May miss custom build processes not in standard files
  • •Generates starting point - user should review and customize
  • •Works best with standard project structures
  • •Does not analyze actual code logic for features

Best Practices

When using this skill:

  1. •Run from project root: Ensure you're in the main project directory
  2. •Review before writing: Check generated content before writing to file
  3. •Customize: Treat output as a template, add project-specific details
  4. •Update regularly: Regenerate when project structure changes significantly
  5. •Backup existing: If README.md exists, back it up first

Error Handling

  • •No project files found: Ask user to confirm working directory
  • •Multiple languages detected: Generate sections for each, note polyglot nature
  • •Existing README: Prompt user before overwriting, offer to merge
  • •Missing key info: Generate placeholder sections with TODO markers

Related Skills

Changelog

Version 1.0.0 (2025-01-13)

  • •Initial release
  • •Support for Python, Node.js, Rust, Go projects
  • •Automatic dependency detection
  • •Standard section generation

Contributing

Found a bug or want to add support for a new project type? Please:

  1. •Open an issue with details
  2. •Submit a PR with improvements
  3. •Follow Contributing Guidelines

License

Apache License 2.0 - See LICENSE

Author

GLINCKER Team