Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
144 lines (98 loc) · 4.31 KB

CONTRIBUTING.md

File metadata and controls

144 lines (98 loc) · 4.31 KB
Raw

Contributing to AI Tools Setup Guide

Thank you for your interest in contributing! This guide helps others replicate your AI development environment setup.

How to Contribute

1. Adding New Tools

To add a new AI tool to the setup guide:

  1. Create a new directory under configs/ (e.g., configs/newtool/)
  2. Add configuration files to that directory
  3. Update README.md with installation and configuration instructions
  4. Update cli.sh to support installing and configuring the new tool

2. Improving Existing Configs

If you want to improve or add features to existing configurations:

  1. Update the relevant config file in configs/<tool>/
  2. Update README.md if the changes affect user workflow
  3. Test changes with ./cli.sh --dry-run first

3. Adding Best Practices

This guide includes software development best practices. To contribute:

  1. Add markdown content to configs/best-practices.md
  2. Reference it in OpenCode configurations if applicable

Submission Guidelines

  1. Keep configurations portable and self-contained
  2. Use relative paths (avoid absolute paths like /Users/username/)
  3. Add comments to explain non-obvious settings
  4. Test the setup script before submitting changes

Style Guide

  • Follows zed-101-setup README format pattern
  • Use emoji headers for visual hierarchy
  • Include copy-paste ready code blocks
  • Keep installation commands simple and verified

Questions?

Open an issue or reach out via support channels listed in README.md.

How to Add New Skills

  1. Create skill directory: Create a new directory under .claude/skills/{skill-name}/
  2. Create SKILL.md file: Add SKILL.md with skill frontmatter and content
  3. Update skill-rules.json: Add triggers and keywords for your new skill
  4. Document: Add skill to skills/README.md if it exists

Skill Frontmatter Format

---
name: skill-name
description: Brief description of what this skill does and when to use it. Include keywords users would mention.
allowed-tools: Read, Grep, Bash(git:*)
model: claude-sonnet-4-20250514
---

# Skill Content

Your skill documentation here...

Best Practices

  • Keep skills focused on a specific domain or pattern
  • Include examples (good and bad patterns)
  • Reference related skills
  • Keep descriptions under 1024 characters

How to Add New Agents

  1. Create agent file: Add .md file to .claude/agents/{agent-name}.md
  2. Define frontmatter: Include name, description, model
  3. Document process: What is agent does and when to invoke it
  4. Create checklist: Organize review criteria by severity (Critical, Warning, Suggestion)
  5. Integration: Reference related skills or commands

Agent Format

---
name: agent-name
description: When to use this agent and what it does.
model: opus
---

# Agent Content

Your agent documentation here...

## Process

Your step-by-step process here...

How to Modify Hooks

  1. Edit settings.json: Add or modify hook definitions
  2. Test hooks: Run hooks manually to verify they work
  3. Update settings.md: Document any new or modified hooks

How to Add New Commands

  1. Create command file: Add .md file to .claude/commands/{command-name}.md
  2. Define frontmatter: Include description, allowed-tools
  3. Document process: Step-by-step instructions
  4. Use variables: $ARGUMENTS, $1, $2, etc.

How to Customize Skill Rules

Edit .claude/hooks/skill-rules.json to:

  • Adjust minConfidenceScore (3-7)
  • Add new skills with triggers
  • Update directory mappings
  • Add exclusion patterns

Testing Changes

Before submitting a PR or commit:

  1. Test skill evaluation: Run /onboard <test> to verify skills activate
  2. Test hooks: Edit a file on main branch to verify branch protection
  3. Test agents: Run /code-reviewer to verify review quality
  4. Test commands: Try new commands in a test directory

Code Style Guidelines

  • Follow your project's existing style
  • Keep documentation concise
  • Use meaningful names for files and functions
  • Avoid over-engineering solutions or overcomplicating things unnecessarily

Remember: Software design is an exercise in human relationships. Make your contributions clear, well-documented, and respectful of existing code.

Questions?

Open an issue or reach out via the support channels listed in README.md.