Privacy-First File Sanitization

About MetadataRemover: Privacy-First File Sanitization

Our mission: Empowering individuals, photographers, and creators to take complete control of their digital footprint with zero server uploads.

Why We Built MetadataRemover

Every modern digital image taken on a smartphone or camera embeds invisible tracking information: precise GPS coordinates of where you live, work, and travel; exact camera models and serial numbers; timestamps down to the millisecond; and hidden software provenance tags.

Most existing online "metadata removal" services require you to upload your sensitive photos and private documents to third-party cloud servers. This introduces a major security risk: your private files can be logged, cached, leaked, or analyzed without your knowledge.

We built MetadataRemover to solve this problem permanently. By leveraging the modern Web Workers and HTML5 File APIs, we moved 100% of the binary parsing and metadata excision directly into your browser memory.

The Zero-Upload Architectural Invariant

Our architecture adheres to a strict Golden Axiom: User files never touch a remote server.

  • Client-Side Binary Parsers: We scan byte streams locally using defensive segment scanners for JPEG (APP1/APP13/COM), PNG (tEXt/zTXt/iTXt chunks), and WebP (RIFF containers).
  • Memory Sandboxing: File operations run in an isolated Web Worker, ensuring high performance without freezing your browser interface.
  • Offline Availability: Once loaded, our Service Worker allows the tool to run completely offline — even in Airplane Mode.
  • Zero Tracking Telemetry: We do not track your file names, contents, or metadata. No external advertising pixels or user analytics scripts inspect your workflow.

Observable Capabilities & Scope Boundaries

We believe authority is built through clear, observable facts rather than inflated claims. Here is exactly what our client-side software does and does not do:

  • Supported File Formats: JPEG/JPG up to 50 MB, PNG up to 50 MB, and WebP up to 35 MB.
  • What Is Stripped: Container-level headers including EXIF camera tags, GPS location coordinates, IPTC copyright fields, XMP packets, PNG text chunks (prompts, seeds, ComfyUI node graphs), and C2PA Content Credentials.
  • Lossless Mode A: Slices container markers at the byte level without recompressing pixels, guaranteeing bit-for-bit identical image quality and preserving ICC color profiles.
  • What Is Not Supported: The engine does not strip invisible pixel watermarks embedded in frequency domains (such as Google SynthID), nor does it read or reset proprietary camera shutter actuation counts stored in vendor MakerNotes.

Engineering Principles & Transparency

We believe in verifiable trust. Anyone can open their browser's Network Inspector (F12) or disconnect their network connection to test and verify that zero file data is transmitted across the wire during inspection or cleaning. We build with lightweight TypeScript and self-hosted assets to ensure rapid page loads and complete privacy.

Have questions, technical feedback, or a bug to report? Reach out via our Contact Page or review our Privacy Policy.