O conteúdo está em inglês; a navegação e os filtros mantêm o idioma selecionado.

Open-Source Tool Claims to Remove Several Types of AI Provenance Marks

A GitHub project called watermarks-remover is designed to handle invisible Unicode characters, file metadata, and text-rewriting techniques intended to disrupt statistical watermark signals. Its documentation lists formats including PNG, JPEG, SVG, PDF, DOCX, HTML, and Markdown, and references provenance systems associated with several AI vendors.

Publicado em 10/09/2026, 04:25:43Última atualização 10/09/2026, 04:25:43Publicação editorial

The project describes several layers of processing: removing invisible Unicode characters, deleting C2PA, EXIF, and XMP metadata, and rewriting text to try to weaken statistical signals. The repository and multiple social posts support the tool’s existence and its stated scope.

The available evidence does not independently establish how effective the tool is against any particular vendor’s watermarking system. It also does not verify the reported 10,000-star milestone or the timing of related disclosures by Anthropic. Rewriting can alter meaning, style, and readability, so the project’s claims should not be treated as independently demonstrated results.

Evidências das fontes 6

A verificação de fontes não garante exatidão permanente. Considere a fonte original, a data e as correções posteriores.

watermarks-remover/skills/remove-ai-marks/SKILL.md at main · guillaumemeyer/watermarks-remover · GitHubgithub.com · Complementar

Raw Copy raw file Download raw file Outline Edit and raw actions | | | --- | | name | remove-ai-marks | | description | Remove multi-vendor AI provenance marks: invisible Unicode (Layer A), statistical text watermarks via rewrite (Layer B, always offer), and C2PA/EXIF/XMP/container metadata on PNG/JPEG/WebP/SVG/PDF/DOCX/ODT/HTML/MD. Covers Claude, Gemini/SynthID-class, OpenAI provenance, and open-LLM sampling marks. Use when the user asks to strip watermarks, remove C2PA/Content Credentials, clean AI metadata, remove invisible Unicode, anti-detect clean AI output, or runs /remove-ai-marks (aliases: /remove-claude-marks). | # Remove AI marks Multi-vendor anti-detection hygiene for text (Unicode + statistical rewrite) and files (C2PA / AI metadata across common containers). [...] ``` "$WM/clean" " $WM "'Content-Type: application/json' ' '"{\"file\": \"$(base64 < shot.png | tr -d '\n')\", \"name\": \"shot.png\", \ " \" \" \"$(base64 < shot.png | tr -d '\n')$(< | '\n' ' ') \" \"