#4848th of 389 in Skills, plugins and prompts
watermarks-remover
Agent skill and HTTP service that strip AI provenance marks from text and files
- Stars
- 27.1k
- License
- MIT
- Last commit
- Nov 2026
- Last release
- Sep 2026
- Language
- Python
98 on 11 Aug 202627,098 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
A Claude Code, Codex, Cursor and Grok skill that calls a stdlib Python HTTP service to remove invisible Unicode, C2PA, EXIF and XMP marks from text, images, documents, audio and video. Statistical text watermarks are handled by an agent rewrite step, with an optional Ollama backend. A PostToolUse hook can check or clean files the agent writes, and a pre-commit gate and CI SARIF export are included.
Who it is for: Coding-agent users who need provenance marks removed from their own content
Strengths
- Core scripts and service need only Python 3.10+ stdlib, no dependencies
- Covers many formats: PNG, JPEG, PDF, DOCX, EPUB, MP4, WAV, MP3, FLAC and more
- PostToolUse hook cleans or reports on every file the agent writes, deterministically
- Installs on Claude Code (plugin marketplace), Codex, Cursor, Cowork and Grok
Weaknesses
- Chat-only text cannot be filtered by a hook; it depends on model-followed skill instructions
- Real PDF stripping requires qpdf; exiftool and c2patool are separate installs
- Statistical watermark removal depends on a rewrite step needing a configured backend
- Skill is a thin client, so cloud sessions need a reachable service URL
What it needs
- Compose
- Needs Python 3.10+, qpdf, exiftool, c2patool, Ollama (optional)
- Models: Ollama (optional rewrite backend)
- port 8765
Quick answers
Is watermarks-remover open source?
Yes. Its code is on GitHub under the MIT license.
Can you run watermarks-remover with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Also in Skills, plugins and prompts
See all 389| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |