watermark-remover
ShadowAqueduct/watermark-remover
This skill removes AI-generated watermarks from text and files, ensuring privacy and hygiene for content you own. It supports multiple file formats and offers both deterministic and statistical watermark removal.
768 73MITUpdated 2026-08-23
Overview
The watermark-remover skill strips AI provenance marks from text and files, covering both invisible Unicode characters and statistical watermarks. It is a thin HTTP client that interacts with a Python service, allowing users to clean text and files without requiring Python on the agent host. The skill supports a wide range of file formats, including text, images, and documents, and provides both inspection and cleaning capabilities.
Capabilities
- ▸Remove invisible Unicode characters and exotic spaces from text
- ▸Strip statistical watermarks from text using agent rewrite and optional scripts
- ▸Clean files in various formats (e.g., PNG, JPEG, PDF, DOCX, HTML, Markdown)
- ▸Provide inspection reports on detected watermarks
- ▸Offer both deterministic and statistical watermark removal methods
Best for
Use this skill to remove AI-generated watermarks from text and files, ensuring that content you own is free from AI provenance marks. It is particularly useful for privacy, content hygiene, and preparing files for public use.
Works with
Claude CodeCursorGitHub CopilotOpenCode
Security considerations
The skill does not execute arbitrary code or handle sensitive data directly. It interacts with a Python service over HTTP, which can be configured with a bearer token for authentication. The service runs in a secure environment and does not store user data.
Categories
Summary and analysis are original content generated by AI Skills Rank. The skill's source text is not reproduced here — view it on the linked repository.