How to Intermediate · 3 min read

How to build code review tool with LLM

Quick answer

Build a code review tool by integrating an LLM like gpt-4o via the OpenAI API to analyze code diffs and provide feedback. Use client.chat.completions.create with prompts that include code snippets and review instructions for actionable suggestions.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0

Setup

Install the OpenAI Python SDK and set your API key as an environment variable for secure access.

bash

pip install openai>=1.0

Step by step

This example shows how to send a code snippet to gpt-4o for review and receive improvement suggestions.

python

import os
from openai import OpenAI

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

code_to_review = '''
def add_numbers(a, b):
    return a + b

print(add_numbers(2, 3))
'''

prompt = f"""You are a senior developer performing a code review. Analyze the following Python code and provide constructive feedback, improvements, and potential bugs.\n\nCode:\n{code_to_review}\n\nReview:"""

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": prompt}]
)

print("Code review feedback:\n", response.choices[0].message.content)

output

Code review feedback:
The function `add_numbers` is simple and correct for adding two numbers. Consider adding type hints for clarity, e.g., `def add_numbers(a: int, b: int) -> int:`. Also, add error handling if inputs might not be numbers. The print statement is fine for testing but should be removed or replaced with proper unit tests in production.

Common variations

Use asynchronous calls with asyncio and await for non-blocking review requests.
Stream responses for real-time feedback display.
Switch models to gpt-4o-mini for cost-effective reviews or claude-sonnet-4-5 for alternative LLMs.

python

import os
import asyncio
from openai import OpenAI

async def async_code_review(code: str):
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
    prompt = f"You are a code reviewer. Review this code:\n{code}\nReview:"
    response = await client.chat.completions.create(
        model="gpt-4o-mini",
        messages=[{"role": "user", "content": prompt}]
    )
    print("Async review feedback:\n", response.choices[0].message.content)

if __name__ == "__main__":
    code_sample = "def foo():\n    pass"
    asyncio.run(async_code_review(code_sample))

output

Async review feedback:
The function `foo` is defined but does nothing. Consider implementing the function or removing it if unused.

Troubleshooting

If you get authentication errors, verify your OPENAI_API_KEY environment variable is set correctly.
For rate limits, implement exponential backoff or switch to a smaller model like gpt-4o-mini.
If code snippets are too long, split them into smaller chunks or summarize before sending.

✅

Key Takeaways

Use gpt-4o with OpenAI SDK v1 for effective code review automation.
Structure prompts to include clear instructions and code context for best feedback.
Async and streaming calls improve user experience in interactive tools.
Handle API limits and large code inputs by chunking or model selection.
Secure API keys via environment variables to protect credentials.

Verified 2026-04 · gpt-4o, gpt-4o-mini, claude-sonnet-4-5

Verify ↗