📘 Code Review Analyzer Tool

🔍 About the Project

Code Review Analyzer is an AI-powered tool that automates the process of analyzing pull requests and code review comments from GitHub.
It fetches merged PRs and their review comments directly from a GitHub repository, and uses OpenAI to extract meaningful insights:
recurring issues, context, and clear recommendations — all grouped and tagged for easy analysis.

---

🚀 1. Clone the Project

git clone https://github.com/OfekSagiv/code-review-analyzer.git
cd code-review-analyzer

---

🛠 2. Install Dependencies

npm install

---

⚙️ 3. Create Environment File

Create a .env file in the root directory and add the following variables:

OPENAI_API_KEY=your_openai_key_here
GITHUB_TOKEN=your_github_token_here
GITHUB_REPO=username/repo-name
DATABASE_URL="postgresql://name:password@localhost:5433/code_reviewer"

Ensure DATABASE_URL matches the Docker setup below: user name, password password, and port 5433

---

🔑 4. Generate API Tokens

🔐 OpenAI API Key:

1. Go to https://platform.openai.com/account/api-keys
2. Click "Create new secret key"
3. Copy and paste it into your .env as OPENAI_API_KEY

🐙 GitHub Token:

1. Go to https://github.com/settings/tokens
2. Click "Generate new token (classic)"
3. Select scopes:
   • repo (for accessing PRs and comments)
4. Copy and paste it into .env as GITHUB_TOKEN

---

🐳 5. Run PostgreSQL with Docker

To run the local database safely, use docker-compose.yml:

✅ This configuration ensures PostgreSQL is accessible only from your machine (localhost) and not exposed to the network or internet.

Start the container:

docker compose up -d

---

🧱 6. Prisma Setup

To connect Prisma to the local PostgreSQL database and create the necessary tables, follow these steps:

1. Generate Prisma client
npx prisma generate

2. Push the schema to the database
npx prisma db push

This will sync the schema defined in prisma/schema.prisma with the local database.

---

🧠 7. Run All Scripts Automatically

This command pulls data from GitHub, tags comments using GPT, and generates an insights report:

node scripts/runAll.js

---

📂 8. Output Files

After running the scripts, you'll get:

• insights_report.txt – a grouped and readable summary of all tagged insights from review comments.

---

✅ Done!

Your insights report is ready – fully sorted, grouped by tags, and clearly documented, making it easy to review patterns and improve code quality. You can use this file as input for chat-based advisors, integrate it into documentation, or build tools that learn from recurring code review insights.