How to Write Your First Prompt

Ted Hisokawa
Aug 12, 2026 19:23

Learn how to effectively use GitHub Copilot’s AI-powered app for coding, with tips on starting prompts, choosing models, and refining tasks.

GitHub has published a guide aimed at helping developers get started with the GitHub Copilot app, an AI-powered assistant designed to streamline coding tasks. The app, part of the broader Copilot ecosystem, enables developers to write, edit, debug, and review code using natural language prompts. This comes as GitHub continues to expand Copilot’s capabilities, with recent updates including larger context windows and configurable reasoning levels.

The guide emphasizes simplicity, encouraging users to describe their tasks in plain English without worrying about perfect syntax. For instance, a prompt like "Add a most-funded sort option to the games list" is enough for Copilot to analyze the codebase and begin implementing changes. The iterative nature of prompting allows users to refine their requests and achieve more precise results over time.

Context Matters

A key feature of the GitHub Copilot app is its ability to work within the context of a connected repository or local project. By linking the app to a GitHub repository or a folder on a local machine, developers provide Copilot access to the necessary files and codebase. This context-driven approach ensures that the AI’s suggestions are relevant and actionable.

Users can also select specific AI models depending on the complexity of the task. While the default model is suitable for most use cases, advanced configurations allow adjustments for more intricate reasoning tasks. Recent updates from June 2026 introduced expanded context windows of up to one million tokens, making Copilot even more effective for handling large-scale projects.

Voice Input and Session Flexibility

For developers who prefer a hands-free experience, the app includes built-in voice input. This feature converts spoken instructions into text, which can then be reviewed and edited before execution. It’s particularly useful for longer or more complex prompts.

Additionally, developers can customize how sessions are run. Options include switching between different agents or enabling remote sessions. This flexibility allows users to pick up their work from any device, maintaining continuity across platforms.

Broader Implications for Developers

GitHub’s focus on agentic development is evident in the Copilot app’s design, which integrates tightly with GitHub artifacts like issues and pull requests. By automating routine coding tasks and providing intelligent suggestions, the app aims to boost productivity while reducing manual effort. With usage-based billing via GitHub AI Credits introduced in June 2026, the app’s accessibility has also expanded, offering plans for individual developers, teams, and educational users.

For developers, this tool represents a significant step forward in integrating AI into daily workflows. As Copilot evolves, its ability to handle larger datasets and more complex reasoning tasks will likely further solidify its role as a core productivity tool in the software development process.

GitHub Universe 2026, scheduled for October 28-29, may offer additional insights into Copilot’s roadmap and new features. Developers interested in exploring the tool can start by downloading the app and experimenting with their first prompt today.

Image source: Shutterstock