Module 03 — Development Workflows

Estimated time: 60–90 minutes

This module covers practical, real-world workflows where GitHub Copilot CLI saves time and reduces friction. Each section focuses on a specific task developers do regularly: reviewing code, debugging, writing tests, and working with Git.

---

Workflow 1 — Code Review from the Terminal

When you want a quick review of a file or function before committing, you can pass code directly to Copilot for feedback.

Reviewing a file

Use cat to pipe a file into Copilot's explain command:

cat auth.py | gh copilot explain

Or use a targeted suggest prompt:

gh copilot suggest "review this Python function for potential security issues: $(cat auth.py)"

What to look for in a code review

When asking Copilot to review code, be specific about what you care about:

• Security: "Are there any security vulnerabilities in this function?"
• Performance: "Could this query be written more efficiently?"
• Readability: "How could this function be made easier to understand?"
• Correctness: "Does this function handle edge cases correctly?"

A vague prompt like "review this code" will produce a general response. Targeted questions produce more actionable feedback.

---

Workflow 2 — Debugging

When you have an error or unexpected behavior, Copilot CLI can help you think through the problem.

Sharing an error message

gh copilot suggest "I'm getting this Python error: ModuleNotFoundError: No module named 'requests'. I have Python 3.11 installed and I'm running the script with python3."

Explaining a stack trace

If you have a stack trace, describe it or paste the key lines:

gh copilot explain "ValueError: invalid literal for int() with base 10: '' at line 42 of my data parser"

Debugging a bash script

When a shell script fails silently or produces unexpected output:

gh copilot suggest "my bash script runs without errors but the output file is empty. I'm using cat file.txt > output.txt in a loop."

Debugging approach

Use Copilot CLI as a thinking partner, not just an answer machine. Ask it to:

1. Explain what the error means
2. List possible causes
3. Suggest which cause is most likely given your situation
4. Propose a fix

This step-by-step approach gives you better results than asking for a direct fix.

---

Workflow 3 — Writing Tests

Generating test cases is one of the most time-saving uses of Copilot CLI.

Generating unit tests

Describe the function you want to test:

gh copilot suggest "write pytest unit tests for a Python function called calculate_discount(price, discount_percent) that returns the discounted price. Test normal cases, zero discount, and 100% discount."

Identifying missing test cases

gh copilot suggest "what edge cases am I missing in tests for a function that parses dates from strings in multiple formats?"

Generating test data

gh copilot suggest "generate 5 sample JSON objects for testing a user registration API. Include valid cases and one case with a missing required field."

Integration test setup

gh copilot suggest "what do I need to set up to run integration tests for a Flask app that connects to a PostgreSQL database?"

---

Workflow 4 — Git Integration

Copilot CLI works well alongside Git to reduce the time spent on common operations.

Writing commit messages

Instead of struggling to write a clear commit message, describe what you changed:

gh copilot suggest "write a commit message for changes that refactor the user authentication module to use JWT instead of session cookies, and add refresh token support"

The output will follow conventional commit format if you ask:

gh copilot suggest "write a conventional commit message for: added input validation to the registration form to prevent SQL injection"

Writing pull request descriptions

gh copilot suggest "write a pull request description for changes that: added pagination to the products API endpoint, set default page size to 20, added page and per_page query parameters"

Understanding a diff

When reviewing a large diff you are unfamiliar with:

gh copilot explain "$(git diff HEAD~1)"

Finding the right Git command

gh copilot suggest "cherry-pick a commit from another branch without committing it yet"
gh copilot suggest "see all commits that touched a specific file"
gh copilot suggest "set up git to use my SSH key for a specific remote"

---

Workflow 5 — Working with Scripts and Automation

Copilot CLI helps you write automation scripts for repetitive tasks.

Creating a deployment script

gh copilot suggest "write a bash script that pulls the latest changes from git, installs npm dependencies, builds the project, and restarts a systemd service called myapp"

Automating file processing

gh copilot suggest "write a Python script that reads all CSV files in a directory, merges them into one file, and removes duplicate rows based on an ID column"

Setting up cron jobs

gh copilot suggest "write a cron expression and the corresponding bash command to run a backup script every day at 2am and log the output to a file"

---

How These Workflows Connect

Each workflow follows a similar pattern: describe the problem or task clearly, let Copilot generate a starting point, then refine based on what you actually need.

The swimlane diagram below shows how the five workflows relate to each other and when to use each one.

---

Exercise — Complete a Full Workflow

Choose one of the following scenarios and work through it using Copilot CLI. Take a screenshot of the most useful Copilot response you receive — you will need this for your submission.

Scenario A — Debugging Find a real error message from a project you are working on (or use a common one like a pip installation failure). Use Copilot CLI to diagnose it and propose a fix.

Scenario B — Test Generation Pick any function you have written recently. Use Copilot CLI to generate unit tests for it. Note how many edge cases it identifies that you had not thought of.

Scenario C — Git Workflow Use Copilot CLI to write a commit message for your most recent commit, or to understand a complex git log or git diff output.

Scenario D — Automation Describe a repetitive task you do manually. Use Copilot CLI to generate a script that automates it.

---

Summary

In this module you:

• Used Copilot CLI to review code for specific concerns
• Approached debugging as a step-by-step conversation with Copilot
• Generated unit tests and identified edge cases
• Used Copilot to improve your Git workflow
• Practised writing automation scripts

Continue to Module 04 — Custom Instructions and Agents.