# Pdf Watermark Remover

> Remove watermarks from PDF files. Use when users request watermark removal, cleaning watermarks, or de-watermarking PDF documents. Supports tiling pattern watermarks (repeated image/text tiles) and semi-transparent text overlay watermarks.

- **Type:** Skill
- **Install:** `agentstack add skill-yuanqi99-manus-skills-pdf-watermark-remover`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [yuanqi99](https://agentstack.voostack.com/s/yuanqi99)
- **Installs:** 0
- **Category:** [Content & Media](https://agentstack.voostack.com/c/content-and-media)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [yuanqi99](https://github.com/yuanqi99)
- **Source:** https://github.com/yuanqi99/manus-skills/tree/main/pdf-watermark-remover

## Install

```sh
agentstack add skill-yuanqi99-manus-skills-pdf-watermark-remover
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# PDF Watermark Remover

Remove watermarks from PDF files using PyMuPDF. Supports tiling pattern watermarks (most common in enterprise document systems) and semi-transparent text overlays.

## Prerequisites

Ensure PyMuPDF is installed:

```bash
sudo pip3 install pymupdf
```

## Usage

Run the bundled script:

```bash
python3 pdf_watermark_remover/cli.py   [--mode auto|pattern|text|both]
```

Or if installed via pip:

```bash
pdf-watermark-remover   [--mode auto|pattern|text|both]
```

### Modes

| Mode | Description |
|---|---|
| `auto` | Detect watermark type automatically and remove (default) |
| `pattern` | Remove only tiling pattern watermarks (repeated tiles covering pages) |
| `text` | Remove only semi-transparent text watermarks |
| `both` | Force remove both types |

### Typical Workflow

1. Receive PDF file from user
2. Run script with `--mode auto` (default)
3. Verify output by viewing 2-3 pages of the result PDF
4. If watermarks remain, retry with `--mode both` or investigate manually
5. Deliver the cleaned PDF to the user

## How It Works

- **Pattern watermarks**: Detected by `/Pattern CS` operators in content streams. Removed by excising `q...Q` graphics state blocks containing pattern fill operations.
- **Text watermarks**: Detected by ExtGState objects with fill opacity (`/ca`) less than 1.0.

## Limitations

- Cannot remove watermarks baked into rasterized images (scanned PDFs)
- Cannot remove watermarks embedded as part of the actual page content (same opacity as normal text)
- For encrypted/password-protected PDFs, decryption must be done first

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [yuanqi99](https://github.com/yuanqi99)
- **Source:** [yuanqi99/manus-skills](https://github.com/yuanqi99/manus-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-yuanqi99-manus-skills-pdf-watermark-remover
- Seller: https://agentstack.voostack.com/s/yuanqi99
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.