Automate PR Descriptions with AI in 10 Minutes

Writing detailed PR descriptions is important for code review, but it's tedious. In this tutorial, you'll build an AI agent that reads your git diff and writes a clear, structured PR description automatically. No more skipped descriptions or vague bullet points.

Step 1: Set up the Python script

Create a file pr_description_generator.py in your repository root.

import os
import subprocess
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

def get_diff(base_branch="main", head_branch=None):
    if head_branch:
        result = subprocess.run(
            ["git", "diff", f"{base_branch}...{head_branch}"],
            capture_output=True, text=True
        )
    else:
        result = subprocess.run(
            ["git", "diff", "origin/main"],
            capture_output=True, text=True
        )
    return result.stdout

def generate_description(diff):
    response = client.chat.completions.create(
        model="gpt-4",
        messages=[
            {"role": "system", "content": "You are a helpful assistant that writes PR descriptions."},
            {"role": "user", "content": f"""
Given the following git diff, write a clear pull request description.
Include:
- Summary of changes (one paragraph)
- Motivation (why these changes were made)
- Key structural changes (new files, major refactors)
- Testing notes (what to verify)

Diff:
```
{diff[:10000]}
```
"""}
        ],
        temperature=0.5,
        max_tokens=500
    )
    return response.choices[0].message.content

if __name__ == "__main__":
    diff = get_diff()
    desc = generate_description(diff)
    print(desc)

Step 2: Create the GitHub Action

Add a workflow file at .github/workflows/pr-description.yml.

name: AI PR Description
on:
  pull_request:
    types: [opened, synchronize]

permissions:
  pull-requests: write
  contents: read

jobs:
  generate:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: '3.10'
      - name: Install dependencies
        run: pip install openai
      - name: Generate PR description
        env:
          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
        id: generate
        run: |
          echo 'DESCRIPTION<> $GITHUB_ENV
          python pr_description_generator.py >> $GITHUB_ENV
          echo 'EOF' >> $GITHUB_ENV
      - name: Update PR body
        uses: actions/github-script@v7
        with:
          script: |
            const body = process.env.DESCRIPTION;
            github.rest.pulls.update({
              owner: context.repo.owner,
              repo: context.repo.repo,
              pull_number: context.issue.number,
              body: body
            });

Step 3: Add your OpenAI API key

Go to your GitHub repository settings → Secrets and variables → Actions. Add a new secret named OPENAI_API_KEY with your key.

Step 4: Try it out

Create a new branch, make some changes, and open a pull request. Within a few seconds, the AI will overwrite the empty PR body with a full description.

Customization Tips

• Use gpt-4-turbo for faster responses.
• Adjust the temperature parameter to control creativity (0.1 for factual, 0.8 for creative).
• Add a base_branch parameter to handle PRs targeting different branches.

That's it! You've built a practical AI agent that saves your team time. Now go ship that feature with a perfect PR description.