Extracted the high-frequency typing animation logic into a new, memoized `TypedText` component.
This prevents the entire `Hero` component (including the heavy `GradientBlinds`) from re-rendering on every character update.
- Created `TypedText` component in `Hero.tsx`
- Wrapped `TypedText` in `React.memo`
- Moved `useTypingEffect` call into `TypedText`
- Updated `Hero` to use `TypedText`
Co-authored-by: ragusa-it <196988693+ragusa-it@users.noreply.github.com>
- Implemented `isValidEmail` utility with strict regex validation (rejects `<` and `>`) to prevent XSS vectors.
- Updated `Contact.tsx` to use `isValidEmail` instead of weak regex.
- Added comprehensive tests for `isValidEmail` in `src/utils/security.test.ts`.
- Fixed flaky test in `src/pages/__tests__/Contact.test.tsx` by clearing `localStorage` in `afterEach`.
- Added test case for invalid email submission.
- Documented findings in `.jules/sentinel.md`.
Co-authored-by: ragusa-it <196988693+ragusa-it@users.noreply.github.com>
- Added `useRateLimit` hook
- Integrated hook into `Contact.tsx`
- Added translations for rate limit error
- Added unit tests
- Fixed type error in `Button.tsx` to allow build to pass
- Replaces the `cancelAnimationFrame` pattern with a boolean ticking flag to reduce function allocation and RAF overhead on high-frequency mousemove events.
- Uses closure variables for coordinates to ensure the latest position is used in the animation frame.
- Improves performance of the custom cursor on high-refresh-rate input devices.
Added strict security headers to `firebase.json` for Firebase Hosting.
Headers included:
- X-Content-Type-Options: nosniff
- X-Frame-Options: DENY
- Referrer-Policy: strict-origin-when-cross-origin
- Content-Security-Policy: Includes directives for 'self', Google Fonts, EmailJS, and disallows object/frame embedding.
Also initialized `.jules/sentinel.md` with the first security learning.
Added `sanitizeInput` utility to escape HTML characters.
Updated `Contact.tsx` to sanitize inputs before sending via `emailjs`.
Added max length validation for Name (100), Subject (200), and Message (5000).
Updated tests to cover sanitization and validation logic, including adding `cleanup()` to prevent test leakage.
