AI Watermark Sweep
A browser-based C2PA metadata and AI-watermark stripper for images, text, and PDFs. Client-side only—all processing happens on your device.
A client-side tool that removes C2PA metadata, JUMBF manifests, and invisible character markers from images (JPEG, PNG, WebP), text files, and PDFs. All processing runs in your browser—nothing is uploaded, no backend, no tracking.
What it does: Strips invisible watermarks and metadata. What it cannot do: Remove word-level watermarks (SynthID, Anthropic) from plain text, because those signals live in the words themselves, not in metadata.
Built with React 19, TypeScript, and Tailwind CSS. Tested with Vitest and Playwright. Design system follows Hallmark anti-AI-slop principles.
ReactTypeScriptTailwind
View experiment →