Git setup helper

This repository includes a small, safe helper to set up your global Git configuration.

Files added:

• .gitconfig.example — template you can copy or inspect.
• scripts/setup-git.sh — a script that shows the commands it would run (dry-run) and can apply them with --apply.

Quick usage:

Dry-run (shows what would be changed):

bash scripts/setup-git.sh --name "Your Name" --email you@example.com

Apply changes:

bash scripts/setup-git.sh --apply --name "Your Name" --email you@example.com

Notes:

• The script defaults to core.editor = code --wait if you have VS Code installed.
• It sets credential.helper=osxkeychain (macOS recommended).
• It will not automatically generate SSH keys; instructions are included in the script output.

Using the GitHub CLI (recommended)

• If you have gh installed and authenticated (gh auth login), this repo includes a simpler gh-backed helper: scripts/setup-git-gh.sh.
• scripts/setup-git-gh.sh uses gh auth setup-git to wire Git to your GitHub credentials and can upload an SSH key with gh ssh-key add.

Dry-run (gh script):

bash scripts/setup-git-gh.sh --name "Your Name" --email you@example.com

Apply (gh script):

bash scripts/setup-git-gh.sh --apply --name "Your Name" --email you@example.com --ssh

Notes:

• scripts/setup-git-gh.sh defaults to dry-run. Use --apply to actually change your global config.
• When you run with --ssh the script will generate an ed25519 key (if missing), add it to the ssh-agent, and upload the public key to GitHub using gh ssh-key add.

If you’d like, I can make the gh script interactive (prompting for missing values), or replace the original scripts/setup-git.sh entirely. Tell me which option you prefer.

Routine Maintenance Script

This script automates the preparation of prompt workflows for VS Code Chat extensions, enabling you to run structured AI-assisted maintenance tasks on your codebase.

Setup

1. Make the script executable:

   chmod +x scripts/routine-maintenance.sh
   

2. Ensure your prompt files are in .github/prompts/ directory

Usage

Basic Commands

# Refactor a specific file
./scripts/routine-maintenance.sh refactor src/main.js

# Generate tests for a file
./scripts/routine-maintenance.sh test src/component.js

# Create documentation for a file or folder
./scripts/routine-maintenance.sh docs src/ --chat

# Debug a problematic file
./scripts/routine-maintenance.sh debug src/buggy.js

# Run full maintenance workflow (refactor → test → docs)
./scripts/routine-maintenance.sh full-maintenance src/feature.js

Advanced Workflows

# Feature development workflow (analyze → design → implement → test → docs)
./scripts/routine-maintenance.sh workflow feature notes/new-feature.md

# Maintenance workflow
./scripts/routine-maintenance.sh workflow maintenance src/legacy.js

# Debug workflow
./scripts/routine-maintenance.sh workflow debug src/error.js

Options

• --chat: Format output for direct paste into VS Code Chat
• --copy: Copy output to clipboard (macOS only)
• --verbose: Show detailed output information

Integration with VS Code Chat

Method 1: Manual (Most Reliable)

1. Run the script with your desired command and --chat --copy flags
2. Open VS Code Chat (Cmd+Shift+I or Cmd+I)
3. Paste (Cmd+V) - the prompt is already in your clipboard
4. The AI will process the prompt and provide the requested output
5. Apply the suggestions back to your codebase

./scripts/routine-maintenance.sh review pages/_posts/my-article.md --chat --copy
# Then paste into VS Code Chat

Method 2: Semi-Automated (VS Code Integration)

Use the vscode-integration.sh script for streamlined workflows:

# Opens file in VS Code with prompt in clipboard
./scripts/vscode-integration.sh vscode-prompt pages/_posts/my-article.md

# Attempts to open VS Code Chat automatically
./scripts/vscode-integration.sh auto-review pages/_posts/my-article.md

Note: VS Code Chat extensions don’t expose a direct CLI API, so full automation requires manual paste step.

Example Workflow

# Prepare refactoring prompt for a component
./scripts/routine-maintenance.sh refactor src/MyComponent.js --chat --copy

# In VS Code Chat, paste the output and get refactoring suggestions
# Apply the changes, then run tests
./scripts/routine-maintenance.sh test src/MyComponent.js --chat --copy

Available Prompts

The script uses these prompt files from .github/prompts/:

• code-refactoring.prompt.md - Code quality improvements
• test-generation.prompt.md - Automated test creation
• documentation.prompt.md - Documentation generation
• debugging.prompt.md - Issue diagnosis and fixes
• requirements-analysis.prompt.md - Requirements extraction
• system-design.prompt.md - Architecture design
• code-implementation.prompt.md - Code writing
• prompt-engineering.prompt.md - Prompt creation

Customization

You can extend the script by:

1. Adding new prompt files to .github/prompts/
2. Creating new workflow types in the cmd_workflow() function
3. Modifying the output formatting in format_for_chat()

Safety Features

• File existence validation
• Large file truncation (shows first 100 lines for files >100KB)
• Error handling for missing prompts
• Safe file content reading

This script transforms your prompt engineering into an executable, repeatable process for maintaining high-quality codebases.

Advanced: Programmatic API Access

While VS Code Chat extensions (GitHub Copilot, etc.) don’t expose a direct CLI API, here are alternatives for background automation:

Option 1: GitHub Copilot CLI (Separate Tool)

npm install -g @githubnext/github-copilot-cli
# Use with your prompts
gh copilot suggest "your prompt here"

Option 2: OpenAI API / Azure OpenAI (Direct)

Create a script that calls the AI model API directly:

# Example using curl with OpenAI API
curl https://api.openai.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -d '{
    "model": "gpt-4",
    "messages": [{"role": "user", "content": "your prompt"}]
  }'

Option 3: VS Code Extension Development

Build a custom VS Code extension that:

1. Reads your .github/prompts/ files
2. Provides commands to trigger workflows
3. Integrates with Chat API programmatically

See the article “Agentic Prompt Workflows in VS Code” for conceptual design.