QueryShield 🛡️

SQL Injection Risk Analysis API + Web UI — A full-stack security service that analyzes SQL query strings, detects suspicious injection patterns, assigns a risk score, and surfaces results through authenticated API endpoints and a React dashboard.

---

What It Does

SQL injection is one of the most common and dangerous vulnerabilities in web applications. QueryShield acts as a pre-execution inspection layer — analyzing SQL query strings before they are executed and returning a structured security risk report.

The system inspects queries for suspicious constructs, assigns weighted scores, and classifies severity levels.

Example attack query caught by QueryShield:

SELECT * FROM users WHERE username = 'admin' OR 1=1 --

QueryShield response:

{
  "id": 1,
  "query_text": "SELECT * FROM users WHERE username = 'admin' OR 1=1 --",
  "risk_score": 60,
  "severity": "high",
  "flags": "OR_ALWAYS_TRUE,COMMENT_INJECTION",
  "analysis_summary": "Multiple indicators consistent with injection behavior",
  "created_at": "2026-03-09T03:39:11.208111",
  "user_id": 1
}

---

Tech Stack

Layer	Technology
Backend	Python, FastAPI
Frontend	React, TypeScript, Vite, Tailwind CSS
Database	PostgreSQL, SQLAlchemy ORM
Auth	JWT (python-jose), Passlib (bcrypt)
Infrastructure	Docker, Docker Compose
Cloud	AWS EC2
API Tooling	Uvicorn, Pydantic

---

Key Features

• JWT Authentication — Secure user registration and login with token-based access control
• SQL Risk Analysis Engine — Detects injection indicators such as always-true conditions, comment truncation, UNION-based attacks, and suspicious schema enumeration
• Risk Scoring System — Assigns numeric scores and severity levels (low, medium, high, critical) based on matched detection rules
• Query Logging — Stores every analyzed query in PostgreSQL, tied to the authenticated user
• Protected Endpoints — History retrieval and high-risk filtering behind auth middleware
• Modern Web Dashboard — React + Tailwind UI that lets you sign up, log in, paste queries, view live analysis, and browse history/high-risk queries
• UX* — Clear navigation, color-coded severity badges, and thoughtful empty/loading/error states that showcase practical product thinking
• Containerized Deployment — Full Docker Compose stack with persistent PostgreSQL volume
• Cloud Hosted — Live on AWS EC2, accessible via public IP

---

API Endpoints

Authentication

Method	Endpoint	Description
POST	/auth/register	Create a new user account
POST	/auth/login	Authenticate and receive a JWT
POST	/auth/token	OAuth2 token endpoint used by Swagger UI

Query Analysis

Method	Endpoint	Description
POST	/query/analyze	Analyze a SQL query and store the result
GET	/query	Retrieve all stored analyses
GET	/query/high-risk	Retrieve only high-risk queries

---

System Architecture

Client Request
      │
      ▼
JWT Authentication Middleware
      │
      ▼
POST /query/analyze
      │
      ▼
SQL Risk Analysis Engine
   ├── Pattern Detection
   ├── Risk Score Calculation
   └── Severity Classification
      │
      ▼
Store Result → PostgreSQL
      │
      ▼
Structured JSON Response → Client

---

Local Setup

# Clone the repo
git clone https://github.com/The1keyy/queryshield.git
cd queryshield

# Set up Python virtual environment
python3 -m venv venv
source venv/bin/activate

# Install backend dependencies
pip install -r requirements.txt

# Run the API (FastAPI + PostgreSQL)
uvicorn app.main:app --reload

Visit the interactive docs at: http://127.0.0.1:8000/docs

---

Web UI (React + Tailwind)

On top of the API, QueryShield ships with a small frontend dashboard non-technical users can try without writing any code.

Screens

• Landing Page — High-level overview with an example analysis and calls-to-action (“Try live analyzer”, “View API docs”)
• Auth — Email + password registration and login, backed by the same JWT auth used by the API
• Live Analyzer — Paste a SQL query, run analysis, and see:
  • Color-coded severity (low, medium, high, critical)
  • Risk score out of 100
  • Matched rule flags (e.g. OR_ALWAYS_TRUE, COMMENT_INJECTION)
  • Human-readable summary of why the query is risky
• History — Table of all your past analyses with timestamps, severities, scores, query preview, and flags
• High-Risk Only — Focused view of queries whose scores exceed the configured high_risk_threshold

Frontend local setup

In a second terminal, from the project root:

cd frontend

# Install frontend dependencies
npm install

# Point the UI at your local API (optional if you keep defaults)
echo 'VITE_API_URL=http://127.0.0.1:8000' > .env

# Run the React dev server
npm run dev

Then open the UI at http://127.0.0.1:5173:

1. Register a new account in the UI (or via /auth/register)
2. Login to receive a JWT stored in the browser
3. Use the Live Analyzer to inspect queries
4. Explore History and High-Risk views to see persisted results

---

Docker Deployment

# Build and start full stack (API + PostgreSQL)
docker compose up -d

This spins up:

• API container running FastAPI on port 8000
• PostgreSQL container with a persistent volume

---

AWS Deployment

QueryShield is live on an AWS EC2 instance:

1. Launch EC2 instance (Ubuntu)
2. Install Docker & Docker Compose
3. Clone repository and build image
4. Run stack with docker compose up -d
5. Open port 8000 via EC2 security group

Access live API docs: http://EC2_PUBLIC_IP:8000/docs

About

Built by Keyshawn Jeannot — Computer Science student at UMass Boston, focused on backend engineering, cloud infrastructure, and defensive security systems.

GitHub · LinkedIn

---

QueryShield demonstrates backend API architecture, secure authentication design, database modeling and persistence, containerized deployment, and cloud infrastructure workflows — applied to a real-world security problem.