Customizing Copilot

Software Development with GitHub Copilot

Thalia Barrera

AI Engineering Curriculum Manager, DataCamp

How custom instructions work

Define:

Coding conventions
Tool preferences
Project context

How custom instructions work

Define:

Coding conventions
Tool preferences
Project context

How to define custom instructions

Include them in: .github/copilot-instructions.md

## Project context
Python project for processing data pipelines with Airflow.


## Coding style
- Use type hints for all functions

- Follow PEP8

- Use snake_case naming

- Write Google-style docstrings


## Testing
- Use pytest

- Include one test per function


## More instructions...

What makes an effective instruction

✅ Effective instructions:

Short and specific
Self-contained
Add context or relevant information

❌ Ineffective instructions:

External documents or repositories
Specific tone or writing style
Limit on detail or formatting

¹ https://docs.github.com/en/copilot/how-tos/configure-custom-instructions/add-repository-instructions

## Style guidelines

- Conform to the coding styles defined 
in styleguide.md in repo my-org/my-repo

- Answer in the style of a friendly 
colleague, using informal language

- Answer in fewer than 1,000 characters

## Style guidelines

- Conform to the coding styles defined 
in styleguide.md in repo my-org/my-repo

- Answer in the style of a friendly 
colleague, using informal language

- Answer in fewer than 1,000 characters

How to enable custom instructions

Save the copilot-instructions.md file
It will be automatically included in your prompts

How to enable custom instructions

Save the copilot-instructions.md file
It will be automatically included in your prompts

Using multiple instruction files

Setting organization-level instructions

Examples of company-wide rules:

Use internal tools
Enforce specific libraries
Apply consistent style
Warn on deprecated patterns

Multi-level instructions priority

Scope	Priority
Personal instructions	Highest
Project instructions	Medium
Organization instructions	Lowest

Example prompt: "Write a function that calculates the average of a list of numbers."

def calculate_average(numbers: List[float]) -> float:
    """Calculate the average of a list of numbers.

    Args:
        numbers (List[float]): A list of numbers to average.

    Returns:
        float: The average value of the numbers.

    Raises:
        ValueError: If the list is empty.
    """
    if not numbers:
        raise ValueError("The list of numbers cannot be empty.")
    return sum(numbers) / len(numbers)

Example prompt: "Write a function that calculates the average of a list of numbers."

def calculate_average(numbers: List[float]) -> float:
    """Calculate the average of a list of numbers.

    Args:
        numbers (List[float]): A list of numbers to average.

    Returns:
        float: The average value of the numbers.

    Raises:
        ValueError: If the list is empty.
    """
    if not numbers:
        raise ValueError("The numbers list cannot be empty.")
    return sum(numbers) / len(numbers)

Let's practice!

Software Development with GitHub Copilot