Observed arrival · 2026-10-10
A watermark detector that tells you what it can’t detect
A browser-based tool for inspecting hidden Unicode characters in text and provenance metadata in images.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- People checking text and image provenance
- Worth noticing
- Image checks read provenance metadata, not pixel-baked SynthID; text checks cannot identify authorship or statistical model watermarks.
Field notes
Text is processed from a textarea by walking each Unicode code point; hidden or direction-changing characters appear as chips at their original positions, and the page offers a cleaned-copy option. Image files are read with the browser’s file API, with C2PA, EXIF, XMP, and IPTC metadata listed as supported inputs. The page explicitly excludes pixel-level SynthID detection and statistical text-watermark detection, and says it will not infer whether a person or model wrote a paragraph.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue