Longer answers for real file tasks.

Remove EXIF, GPS, photo metadata, and optional AI Content Credentials from one photo or a folder locally. Inspect first, then save verified clean copies.

No. The decoder and image encoder run in this browser. The page does not send the selected file to NullKit or a third-party conversion service.

No. PNG prevents further lossy compression, but it cannot restore detail already removed by JPEG compression.

They are replaced with the background color you select before conversion.

Not always. The result depends on the image and selected quality, so NullKit reports the exact size before download.

Yes. Select image-watermark mode and choose a local PNG, JPG, or WebP logo.

No. Decoding, resizing, encoding, and size comparison happen in this browser without a file-upload request.

Yes. Proportions remain locked by default. You may unlock them deliberately when an exact width and height are required.

It keeps a normal SDR JPEG as the fallback, adds a smaller JPEG gain map, and includes the metadata and MPF index that compatible HDR viewers use to show brighter highlights.

For this image tool, the selected pixels are replaced in the exported bitmap rather than covered by a removable document annotation. You should still inspect the final copy carefully.

No. It removes supported document properties only. Visible text and page content require a dedicated redaction workflow.

Yes, when the browser decodes the source transparency correctly. The preview uses a checkerboard so transparent areas remain visible.

PNG is lossless and often needs more bytes than compressed WebP, especially for photographs.

No. Payload construction and QR rendering happen in the current browser tab.

No. The tool displays the exact decoded content first and requires a separate deliberate action before opening an HTTP or HTTPS link.

The rules include common analytics, advertising, click, and social-sharing identifiers such as utm_* values, fbclid, gclid, mc_cid, and several platform-specific sharing tags.

No. Pixel adjustment, preview generation, and export happen in this browser without a file-upload request.

WebP can contain transparent pixels but JPG cannot. The selected color is placed behind the image before JPG encoding.

Yes. The selection surface uses pointer input and works with touch, pen, or mouse in supported browsers.

No. Never enter a recovery phrase, password, backup code, encryption key, or similar secret into this checker.

It produces a readiness level and prioritized actions, not a scientific or universal privacy score. The recommendations are more useful than a dramatic percentage.

No. The selected files are read and copied into a new PDF in this browser.

Yes. Define up to 100 named page groups or split the source into fixed-size parts, then download one ZIP archive.

No. The browser reads the bytes and uses its built-in cryptographic digest function locally.

No. Generation uses crypto.getRandomValues in this browser and the page does not submit the result.

No. Pattern detection and replacement run in the current tab without a submission request.

No. The model and inference runtime load as static NullKit assets, and processing occurs on this device.

Yes. The page cards and movement controls are designed for touch, although large documents may work better on a computer.

No. Cropping changes the visible page boundary and is not redaction or sanitization.

Yes. Selected pages are placed onto new pages using the target dimensions and fit mode you choose.

Yes. Each selected image becomes one PDF page in the displayed order.

No. ZIP parsing, PDF analysis, image processing, hashing, and output verification run in this browser tab.

No. Unlocking requires the correct user or owner password and does not attempt guessing or circumvention.

No. It can reconstruct recoverable structure, but bytes that are absent or irreparably damaged cannot be recreated.

Not in lossless mode. Target-size mode can reduce resolution and JPEG quality after you explicitly approve rendered output.

Use 144 DPI for ordinary screen viewing and 216 or 300 DPI only when extra detail is genuinely needed.

No. PDF to images renders whole pages; this tool attempts to recover image objects used inside pages.

Not exactly. The tool prioritizes readable text and uses heuristics because PDFs do not consistently encode semantic structure.

No. Raw HTML is treated as text to keep rendering predictable and avoid unsafe markup.

No. It is a visible deterrent or label and can be altered by sufficiently capable software.

Learn how to remove selected visible information from a PDF with a flattened local export, then verify the redacted copy before sharing it.

No. Field values remain in the current browser tab and are written into the downloaded copy locally.

No. It can show differences between two supplied files but cannot establish which one is genuine.

Not currently. Existing-text replacement is visibly disabled until replacement text can be composited only after the source pixels are irreversibly masked. You can still add and edit new text boxes.

No. The OCR worker, English language data, and PDF processing assets are served by NullKit and run in the current browser.