Spring Boot Security Review
Use when adding auth, handling input, creating endpoints, or dealing with secrets.
When to Activate
- Adding authentication (JWT, OAuth2, session-based)
- Implementing authorization (@PreAuthorize, role-based access)
- Validating user input (Bean Validation, custom validators)
- Configuring CORS, CSRF, or security headers
- Managing secrets (Vault, environment variables)
- Adding rate limiting or brute-force protection
- Scanning dependencies for CVEs
Authentication
- Prefer stateless JWT or opaque tokens with revocation list
- Use
httpOnly,Secure,SameSite=Strictcookies for sessions - Validate tokens with
OncePerRequestFilteror resource server
Authorization
- Enable method security:
@EnableMethodSecurity - Use
@PreAuthorize("hasRole('ADMIN')")or@PreAuthorize("@authz.canEdit(#id)") - Deny by default; expose only required scopes
Input Validation
- Use Bean Validation with
@Validon controllers - Apply constraints on DTOs:
@NotBlank,@Email,@Size, custom validators - Sanitize any HTML with a whitelist before rendering
SQL Injection Prevention
- Use Spring Data repositories or parameterized queries
- For native queries, use
:parambindings; never concatenate strings
Password Encoding
- Always hash passwords with BCrypt or Argon2 — never store plaintext
- Use
PasswordEncoderbean, not manual hashing
CSRF Protection
- For browser session apps, keep CSRF enabled; include token in forms/headers
- For pure APIs with Bearer tokens, disable CSRF and rely on stateless auth
Secrets Management
- No secrets in source; load from env or vault
- Keep
application.ymlfree of credentials; use placeholders - Rotate tokens and DB credentials regularly
Security Headers
CORS Configuration
- Configure CORS at the security filter level, not per-controller
- Restrict allowed origins — never use
*in production
Rate Limiting
- Apply Bucket4j or gateway-level limits on expensive endpoints
- Log and alert on bursts; return 429 with retry hints
Dependency Security
- Run OWASP Dependency Check / Snyk in CI
- Keep Spring Boot and Spring Security on supported versions
- Fail builds on known CVEs
Logging and PII
- Never log secrets, tokens, passwords, or full PAN data
- Redact sensitive fields; use structured JSON logging
File Uploads
- Validate size, content type, and extension
- Store outside web root; scan if required
Checklist Before Release
- Auth tokens validated and expired correctly
- Authorization guards on every sensitive path
- All inputs validated and sanitized
- No string-concatenated SQL
- CSRF posture correct for app type
- Secrets externalized; none committed
- Security headers configured
- Rate limiting on APIs
- Dependencies scanned and up to date
- Logs free of sensitive data
Remember: Deny by default, validate inputs, least privilege, and secure-by-configuration first.


