Hugging Face's logo

Spaces:
Padmanav
/
ai-code-review-agent
Running

App Files Community
ai-code-review-agent / README.md
Padmanav's picture
Padmanav
fix Hugging Face README YAML metadata
3472803
preview code
|
Raw
History Blame Contribute Delete
1.8 kB

A newer version of the Gradio SDK is available: 6.20.0

Upgrade
metadata 
title: AI Code Review Agent
emoji: 🤖
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.31.0
app_file: gradio_app/app.py
pinned: false

🤖 AI Code Review Agent

CI codecov Python License

A multi-agent AI platform that simulates a software engineering team reviewing your code.

What it does

  1. Repository Analysis — Detects language, frameworks, architecture
  2. Bug Detection — Runs Ruff + Bandit static analysis, then LLM review
  3. Test Generation — Creates pytest test cases for uncovered functions
  4. Code Review — Reviews SOLID principles and suggests refactors
  5. Full Report — Combines everything into a structured engineering report

Setup

This Space requires one secret:

Secret Description
OPENROUTER_API_KEY Your OpenRouter API key

Go to Space Settings → Secrets and add OPENROUTER_API_KEY.

Local development 

git clone https://github.com/Padmanav-Mohanty/ai-code-review-agent.git
cd ai-code-review-agent
pip install -r requirements.txt
export OPENROUTER_API_KEY=your_key_here
python gradio_app/app.py

Tech Stack

  • Agents: Custom multi-agent pipeline (repo analysis, bug detection, test gen, code review)
  • LLM: OpenRouter (default: meta-llama/llama-3.3-70b-instruct)
  • Frontend: Gradio
  • Static Analysis: Ruff, Bandit
  • Repo Cloning: GitPython