🤖 Codentify – AI Code Reviewer

Codentify is a smart AI-powered web app that reviews, analyzes, and improves your code using LLMs. A powerful and customizable AI-powered code reviewer web application built using Python, Streamlit, and OpenRouter’s free LLMs (like Mistral-7B).

This app allows developers to paste or upload source code, select review focus and explanation style, and instantly receive a professional AI-generated review along with corrected code.

📸 App View

Web View	APP's Specification

🚀 Features

• 🔍 AI-Powered Code Review: Analyze logic, bugs, performance, and security flaws
• 📥 Upload or Paste Code: Supports .py, .js, .jsx, .java, .ts, .cpp, .cs
• 🧠 Select Review Focus:
  • General Review
  • Bug Detection
  • Performance Optimization
  • Security Issues
  • Code Refactoring
• 💬 Choose Explanation Style:
  • Detailed
  • Concise
  • Line-by-Line
  • Bullet Points
• 🌐 Supports Multiple Languages:
  • Python, JavaScript, Java, C++, C#, Go, TypeScript
• ✨ Dynamic Prompt Engineering for maximum accuracy
• 📘 Project Overview Tab: Includes specs, tech stack, and documentation
• 🎨 Professional UI: Responsive layout, dark theme, clean footer with profile links

---

🛠 Tech Stack

Layer	Tool / Tech
Frontend	Streamlit
Backend Logic	Python
AI Model	Mistral-7B (via OpenRouter)
Prompting	JSON-powered dynamic prompt template
Styling	Custom CSS (style.css)
Secrets Mgmt	.env + python-dotenv

---

📂 File Structure

├── main.py # Main Streamlit app
├── prompt_template.json # Dynamic LLM prompt template
├── style.css # Custom styles and footer
├── .env # OpenRouter API key (NOT shared)
├── requirements.txt # Install dependencies
└── README.md # You're here

---

⚙️ Installation & Run

1. Clone the Repository

git clone https://github.com/AritraOfficial/Codentify.git
cd Codentify

2. Set Up the Environment

• Create .env with your OpenRouter API key:

OPENROUTER_API_KEY= sk-xxxxxx-your-key

3. Install Dependencies

pip install -r requirements.txt

4. Launch the App

streamlit run main.py

---

📸 Computer's View

Upload File	Ask Questions	Get Answers

---

---

📸 Mobile's View

APP View	Footer View	APP's Specification

---

• 🔐 Free LLMs Used

 This app uses Mistral-7B Instruct from OpenRouter, a fast and free-to-use model for code understanding and natural language generation.
You can switch to other free OpenRouter models by changing the model name in main.py.

🧠 How It Works

1. User selects:

   • ✅ Language preference
   • 🔍 Type of review
   • 🧠 Style of explanation

2. User pastes or uploads code

3. Code is validated against the selected language

4. A dynamic prompt is constructed using all selections

5. The LLM returns:

   • ⚠️ Problem analysis
   • 💡 Suggestions for improvement
   • 🛠️ Corrected version of the code

---

✨ Example Output

🔹 Issues Found:
- Missing error handling
- Unused variable

🔹 Suggestions:
- Add try-except block
- Remove unused variable

✅ Corrected Code:
```python
def process(data):
    try:
        return data.lower()
    except Exception:
        return None

---

👤 Developer

Aritra Mukherjee - AI Developer

---

📝 License

This project is licensed under the MIT License.

---

🤝 Contributing

Feel free to:

• Report bugs
• Suggest features
• Fork and improve the app!

Open a PR or contact me via LinkedIn or email.

---

📌 Disclaimer

This tool is powered by LLMs and should not be used for production-critical code reviews without human verification.

---

🙌 Acknowledgments

• Streamlit
• OpenRouter
• Mistral

---

📧 Contact

For queries or collaborations, feel free to connect: