Watermarks Remover

watermarks-remover

A plugin that strips hidden AI provenance marks from text, images, and documents: invisible Unicode, C2PA, EXIF, XMP

Plugin

Medium risk

We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.

Why this level

  • Modifies the user's files, including metadata and text
  • Needs a separately running local HTTP service for full functionality
All reasons and checks

guillaumemeyer/watermarks-remover

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add plugins/watermarks-remover

Detects the agents on your machine, checks the risk and pins the version.

Other ways to install

Run one by one in the Claude Code chat

/plugin marketplace add guillaumemeyer/watermarks-remover
/plugin install watermarks-remover@watermarks-remover

Checked against the repository on Sep 25, 2026, commit 6cbc0ae.

Text for your agent

In Claude Code, add the marketplace and install the plugin: /plugin marketplace add guillaumemeyer/watermarks-remover, then /plugin install watermarks-remover. For other hosts run python3 install_skill.py --skill remove-ai-marks --target cursor (or claude-code, cowork).

Other ways from the author
/plugin marketplace add guillaumemeyer/watermarks-remover
/plugin install watermarks-remover

This is third-party code. Review the repository files before installing.

What it does

The plugin bundles two skills. remove-ai-marks strips three classes of marks: invisible Unicode characters and unusual spacing, statistical text watermarks by having the agent rewrite the passage, and provenance metadata such as C2PA, EXIF, XMP, or document properties on images, PDFs, DOCX, XLSX, PPTX, video, and audio. It works as a thin client to a local Python HTTP service and runs no code itself. The second skill, clean-user-facing-text, is self-contained and cleans only text without an external service. The plugin can also register a PostToolUse hook that automatically checks or immediately cleans every file the agent writes.

Who it is for. For authors and editors who want content hygiene on their own material: stripping hidden marks before publishing something they wrote or generated themselves.

Good fit when

  • You need to check your own text or file for hidden provenance marks before publishing
  • You want the agent to automatically clean every file it writes
  • You need to strip C2PA, EXIF, or XMP metadata from an image or document you own

Not a fit when

  • The content does not belong to the user or was not created by them
  • The plugin's local HTTP service is not and will not be running for the remove-ai-marks skill

Example request

Check this file for hidden AI watermarks and clean it if you find any

Limitations

The remove-ai-marks skill carries no code of its own and needs a separately running local Python service, whose address can be overridden with WATERMARKS_SERVICE_URL. The tool is meant only for content the user owns; the author is explicit about this as a matter of responsible use.

How to disable. Remove the plugin via /plugin in Claude Code, or delete the skill folders from ~/.claude/skills or ~/.cursor/skills.

Security check

  • Modifies the user's files, including metadata and text
  • Needs a separately running local HTTP service for full functionality

README in short

The README breaks marks into three layers: invisible Unicode, statistical text watermarks, and file metadata, with a table of supported formats from PNG to EPUB. For Claude Code the fastest path is the plugin marketplace; other hosts use install_skill.py with a target flag: claude-code, claude-project, cowork, or cursor. It separately covers the PostToolUse hook for automatically checking written files, and notes that the installer validates the skill against the Agent Skills specification. MIT licensed.

FAQ

Does Python need to run on the agent's side?

The remove-ai-marks skill itself carries no code and talks to the service over HTTP, but the Python service needs to run somewhere separately.

Can it automatically clean every written file?

Yes, the plugin registers a PostToolUse hook with check and clean modes, set through the plugin's settings or the WATERMARKS_HOOK_MODE variable.

Editors’ pick

A skill that strips AI writing tells from text using Wikipedia's list, without adding invented facts

SkillLow riskNo VPN needed52.7KRepository stars

Marketing Skills for AI agents

Marketing Skills for AI Agents

Editors’ pick

About fifty skills for CRO, copywriting, SEO, analytics, ads and launches, built for coding agents

SkillLow risk51.8KRepository stars
Official

SendPulse's official remote MCP server: statistics, campaigns and user data through an agent chat

MCP serverHigh riskRussian stack1Repository stars
Editors’ pick

An MCP server and CLI covering all 113 Yandex Direct API v5 methods generated from WSDL, exposing 9 read tools by default and writes only on demand

MCP serverHigh riskRussian stackNo VPN needed1Repository stars
Foxx AIWatermarks Remover

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.