README.md

metadata

title: SQL Analyzer
emoji: 🔍
colorFrom: blue
colorTo: indigo
sdk: docker
pinned: false
short_description: 'SQL analyzer: linter, AST explorer, formatter & detector'

SQL Analyzer — Linter, AST Explorer & Injection Detector

An interactive SQL analysis tool powered by SQLFluff 4.x, built as a pure Python FastAPI backend serving a React frontend.

Features

• SQL Linter — SQLFluff rule violations with severity, rule codes, line/column info, and fixable indicators
• AST Tree Explorer — Interactive collapsible/expandable parse tree with search, filter, and color-coded node types
• SQL Injection Detector — Detects tautologies, stacked queries, UNION exfiltration, comment bypasses, and more
• SQL Formatter — Auto-fix and format SQL with copy-to-clipboard and apply-to-editor
• Swagger UI — Full OpenAPI 3.1 documentation at /swagger
• 17 SQL dialects — ANSI, PostgreSQL, MySQL, T-SQL, SQLite, BigQuery, Snowflake, Redshift, DuckDB, Hive, Spark SQL, Trino, Databricks, Oracle, Teradata, ClickHouse, Athena

Architecture

sql-analyzer-standalone/
├── api/
│   ├── main.py          ← FastAPI app (REST endpoints + static file serving)
│   └── static/          ← Built React bundle (generated by pnpm build)
├── frontend/
│   ├── src/             ← React 19 + TypeScript source
│   ├── package.json
│   └── vite.config.ts   ← Builds into ../api/static/
├── requirements.txt
├── Dockerfile
└── README.md

Local Development

# 1. Install Python deps
pip install -r requirements.txt

# 2. Build the React frontend
cd frontend && pnpm install && pnpm build && cd ..

# 3. Start the FastAPI server
cd api && uvicorn main:app --reload --port 7860

Open http://localhost:7860

API Endpoints

Method	Path	Description
GET	/api/health	Health check + SQLFluff version
POST	/api/lint	Lint SQL with SQLFluff
POST	/api/parse	Parse SQL into AST
POST	/api/format	Format/fix SQL
POST	/api/inject	Detect SQL injection patterns
GET	/openapi.json	OpenAPI schema
GET	/docs	Swagger UI (FastAPI built-in)
GET	/swagger	Custom Swagger UI page

Request Format

All POST endpoints accept:

{
  "sql": "SELECT * FROM users WHERE id = 1",
  "dialect": "ansi"
}

Supported dialects: ansi, postgres, mysql, tsql, sqlite, bigquery, snowflake, redshift, duckdb, hive, sparksql, trino, databricks, oracle, teradata, clickhouse, athena