🚀 AI Dev Tasks for GitHub Copilot 🤖

Welcome to AI Dev Tasks! This repository provides a collection of .prompt.md files designed to supercharge your feature development workflow within GitHub Copilot Chat. By leveraging these reusable prompt files, you can systematically approach building features, from ideation to implementation, with built-in checkpoints for verification.

Stop wrestling with monolithic AI requests and start guiding your AI collaborator step-by-step!

✨ The Core Idea

Building complex features with AI can sometimes feel like a black box. This workflow aims to bring structure, clarity, and control to the process by:

1. Defining Scope: Clearly outlining what needs to be built with a Product Requirement Document (PRD).
2. Detailed Planning: Breaking down the PRD into a granular, actionable task list.
3. Iterative Implementation: Guiding the AI to tackle one task at a time, allowing you to review and approve each change.

This structured approach helps ensure the AI stays on track, makes it easier to debug issues, and gives you confidence in the generated code.

Workflow: From Idea to Implemented Feature 💡➡️💻

Here's the step-by-step process using the .prompt.md files in this repository:

1️⃣ Create a Product Requirement Document (PRD)

First, lay out the blueprint for your feature. A PRD clarifies what you're building, for whom, and why.

You can create a lightweight PRD directly within GitHub Copilot Chat:

1. Ensure you have the prompt files from this repository in your .github/prompts folder.

2. In GitHub Copilot Chat, initiate PRD creation using the /create-prd command:

   /create-prd
   Here's the feature I want to build: [Describe your feature in detail]
   

   The AI will ask clarifying questions to help create a comprehensive PRD. Answer these questions to provide the necessary context for your feature.

2️⃣ Generate Your Task List from the PRD

With your PRD drafted (e.g., prd-MyFeature.md), the next step is to generate a detailed, step-by-step implementation plan.

1. In GitHub Copilot Chat, use the /generate-tasks command:

   /generate-tasks
   Use the PRD file: prd-MyFeature.md
   

   The AI will first generate high-level tasks and ask for confirmation before creating detailed sub-tasks.

3️⃣ Examine Your Task List

You'll now have a well-structured task list, often with tasks and sub-tasks, ready for implementation. This provides a clear roadmap for development.

4️⃣ Start Working Through Tasks

To ensure methodical progress and allow for verification, use the /process-task-list command. This instructs the AI to focus on one task at a time and wait for your approval before moving to the next.

1. In GitHub Copilot Chat, start processing your task list:

   /process-task-list
   Start with the first sub-task in: tasks-MyFeature.md
   

   The AI will work on the first task and then prompt you to review before continuing.

5️⃣ Review, Approve, and Progress ✅

As the AI completes each task, you review the changes:

• If the changes are good, simply reply with "yes" or "y" to instruct the AI to mark the task complete and move to the next one.
• If changes are needed, provide feedback to the AI to correct the current task before moving on.

You'll see a satisfying list of completed items grow, providing a clear visual of your feature coming to life!

While it's not always perfect, this method has proven to be a very reliable way to build out larger features with AI assistance.

About GitHub Copilot Prompt Files

This workflow leverages GitHub Copilot's prompt files feature, which allows you to create reusable prompts stored as .prompt.md files in your workspace. These files provide:

• Structured prompts: Consistent formatting and behavior across your team
• Reusability: Save time by reusing proven prompt patterns
• Team collaboration: Share prompts across your team through version control
• Built-in modes: Support for different modes (ask, edit, agent) and tool integration

Learn more about prompt files in the GitHub Copilot documentation.

🗂️ Files in this Repository

• create-prd.prompt.md: Guides GitHub Copilot in generating a comprehensive Product Requirement Document for your feature through interactive questioning.
• generate-tasks.prompt.md: Takes a PRD file as input and helps GitHub Copilot break it down into a detailed, step-by-step implementation task list.
• process-task-list.prompt.md: Instructs GitHub Copilot on how to process the generated task list, tackling one task at a time and waiting for your approval before proceeding.

These prompt files are located in the .github/prompts/ directory and can be invoked in GitHub Copilot Chat using the / command syntax (e.g., /create-prd).

🌟 Benefits

• Structured Development: Enforces a clear process from idea to code.
• Step-by-Step Verification: Allows you to review and approve AI-generated code at each small step, ensuring quality and control.
• Manages Complexity: Breaks down large features into smaller, digestible tasks for the AI, reducing the chance of it getting lost or generating overly complex, incorrect code.
• Improved Reliability: Offers a more dependable approach to leveraging AI for significant development work compared to single, large prompts.
• Clear Progress Tracking: Provides a visual representation of completed tasks, making it easy to see how much has been done and what's next.

🛠️ How to Use

1. Setup Prompt Files: Copy the .prompt.md files from this repository to your project's .github/prompts/ directory.
2. Enable Prompt Files: Ensure that GitHub Copilot prompt files are enabled in your VS Code settings.
3. Follow the Workflow: Systematically use the prompt commands in GitHub Copilot Chat as described in the 5-step workflow above:
   • /create-prd to generate your PRD
   • /generate-tasks to break down the PRD into tasks
   • /process-task-list to implement tasks step-by-step
4. Adapt and Iterate:
   • Feel free to modify the prompts within the .prompt.md files to better suit your specific needs or coding style.
   • If the AI struggles with a task, try rephrasing your initial feature description or breaking down tasks even further.

💡 Tips for Success

• Be Specific: The more context and clear instructions you provide (both in your initial feature description and any clarifications), the better the AI's output will be.
• Answer Clarifying Questions: The /create-prd prompt is designed to ask questions to gather comprehensive requirements. Take time to provide detailed answers.
• Review Each Step: The step-by-step approach allows you to catch issues early and provide course corrections.
• Use GitHub Copilot Agent Mode: The prompt files are configured to use agent mode, which provides access to codebase context and other helpful tools.
• Patience and Iteration: AI is a powerful tool, but it's not magic. Be prepared to guide, correct, and iterate. This workflow is designed to make that iteration process smoother.

🙏 Attribution

This project is based on the original work from snarktank/ai-dev-tasks. Special thanks to the original creators for developing this structured approach to AI-assisted development with GitHub Copilot.

🤝 Contributing

Got ideas to improve these .prompt.md files or have new ones that fit this workflow? Contributions are welcome! Please feel free to:

• Open an issue to discuss changes or suggest new prompt files.
• Submit a pull request with your enhancements.

---

Happy AI-assisted developing with GitHub Copilot! 🚀