Auto SQL Reviewer & Formatter

A simple web-based tool to review and format SQL queries using a local Large Language Model (LLM) via Ollama. It provides two main features:

• Review SQL → Detect performance issues, anti-patterns, and suggest improvements.
• Format SQL → Return clean, properly indented SQL with standard casing.

---

🚀 Features

• Uses a local LLM model (default: llama3) for processing.
• REST API backend written in Go.
• Simple HTML/JS frontend.
• Works entirely offline once the model is downloaded.
• No streaming — faster to return a complete response.

---

📂 Project Structure

├── cmd
│   └── server
│       └── main.go
├── go.mod
├── go.sum
├── internal
│   ├── api
│   │   ├── format.go
│   │   └── review.go
│   └── llm
│       ├── localllm.go
│       ├── openai.go
│       └── provider.go
├── readme.md
└── static
    └── index.html

---

🛠 Prerequisites

• Go ≥ 1.20
• Docker (if running Ollama inside Docker)
• Ollama installed locally or running via Docker
• A downloaded model (default: llama3)

---

⚙️ Setup Instructions

1️⃣ Clone the repository

git clone https://github.com/yourusername/auto-sql-reviewer.git
cd auto-sql-reviewer

2️⃣ Install dependencies

go mod tidy

3️⃣ Run Ollama

Option A — Native installation

Install Ollama from ollama.ai/download, then:

ollama pull llama3
ollama serve

Option B — Docker

docker run -d --name ollama \
  -p 11434:11434 \
  -v ollama:/root/.ollama \
  ollama/ollama

docker exec -it ollama ollama pull llama3

---

4️⃣ Environment variables

You can override the model via:

export LOCAL_LLM_MODEL=llama3

(Default: llama3)

---

5️⃣ Run the Go API server

go run main.go

This will start the API on http://127.0.0.1:8080.

---

6️⃣ Open the Web UI

Open index.html in your browser.

• Enter SQL in the textarea
• Click Review SQL or Format SQL
• See results in the separate output block

---

📡 API Endpoints

Review SQL

curl -X POST http://127.0.0.1:8080/review \
  -H "Content-Type: application/json" \
  -d '{"sql":"SELECT * FROM orders;"}'

Format SQL

curl -X POST http://127.0.0.1:8080/format \
  -H "Content-Type: application/json" \
  -d '{"sql":"select * from orders;"}'

---

⚠️ Common Issues

Issue	Cause	Fix
model "llama3" not found	Model not pulled	Run ollama pull llama3 inside container or locally
Slow responses	Large models take time to process	Use smaller model (llama2, mistral, etc.)
permission denied /var/run/docker.sock	Docker socket access issue	Add user to docker group: sudo usermod -aG docker $USER

---

📜 License

MIT License. Do whatever you want, just give credit.