This MCP server removes the "Made with Gamma" watermark from PDF and PowerPoint (PPTX) files exported from Gamma.app, structurally and losslessly — no re-rendering, no uploading.
Key capabilities:
Remove Gamma watermarks (remove_gamma_watermark): Deletes the badge image and its gamma.app hyperlink from .pdf or .pptx files. For PDFs, it drops link annotations and the badge image draw operation; for PPTX files, it removes the gamma-hyperlinked shape from slide masters, layouts, and slides. Reports exactly what was removed and warns if a watermark may remain (e.g., flattened into a page image).
Check file compatibility (check_gamma_watermark_support): Verifies whether a file is a supported format before processing, with guidance if not.
Custom output path: Optionally specify an output path; defaults to <name>-no-watermark.<ext> next to the input file.
Flexible deployment: Run locally via uvx/pipx, integrate with Claude Desktop or Claude Code, or access via a remote streamable-HTTP endpoint for agents that can't run a local server.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@gamma-watermark-remover-mcp remove the gamma watermark from ~/Downloads/presentation.pptx"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP server that lets Claude Desktop, Claude Code, and any MCP client remove the "Made with Gamma" watermark from PDF and PowerPoint (.pptx) files exported from Gamma.app — right on your machine. Just ask:
"remove the gamma watermark from ~/Downloads/deck.pdf"
Removal is structural and lossless: the badge image and its gamma.app hyperlink are deleted as document objects. Nothing is re-rendered, nothing is uploaded.
The Gamma Watermark Remover MCP server connects the Model Context Protocol to a document-cleaning engine, so AI assistants can remove the "Made with Gamma" watermark from Gamma.app exports on your behalf. Instead of opening a separate tool, you stay in your conversation: point Claude at an exported PDF or PowerPoint file, and the badge — a discrete document object with a gamma.app hyperlink — is deleted structurally. Your text stays selectable, your slides stay editable, and with the local server nothing ever leaves your machine.
This is the same engine behind the gammaremover.com web app and the gamma-watermark-remover CLI, packaged for the MCP ecosystem: Claude Desktop, Claude Code, Cursor, Windsurf, and any other MCP-compatible client.
Related MCP server: PDF MCP Flow
Add to claude_desktop_config.json (Settings → Developer → Edit Config):
{
"mcpServers": {
"gamma-watermark-remover": {
"command": "uvx",
"args": ["gamma-watermark-remover-mcp"]
}
}
}claude mcp add gamma-watermark-remover -- uvx gamma-watermark-remover-mcpThe server speaks stdio. Command: uvx gamma-watermark-remover-mcp (or pipx run gamma-watermark-remover-mcp, or pip install gamma-watermark-remover-mcp then gamma-watermark-remover-mcp).
A hosted streamable-HTTP endpoint is available at https://gammaremover.com/mcp for agents that can't run a local server. It accepts file content as base64 (≤30MB), processes in memory only — nothing is stored or logged. For maximum privacy prefer the local stdio server above or the in-browser tool.
Tool | What it does |
| Cleans a |
| Checks whether a file is a supported format before processing |
If the badge was flattened into the page image (rare), the result honestly says a watermark may remain instead of silently failing.
This server wraps the gamma-watermark-remover Python library:
PDF (pypdf): drops link annotations targeting gamma.app/gamma.to and the draw op of the small bottom-right badge image (transformation-matrix tracked, size-guarded so backgrounds are never touched)
PPTX (python-pptx): removes the gamma-hyperlinked shape from slide masters/layouts (where Gamma stores it) and slides
Which MCP clients does this work with?
Any client that speaks MCP over stdio or streamable HTTP: Claude Desktop, Claude Code, Cursor, Windsurf, Cline, and others. The stdio variant is the default; the hosted endpoint at https://gammaremover.com/mcp covers clients that cannot spawn local processes.
Do my files get uploaded? With the local stdio server, never — processing happens on your machine. The hosted endpoint receives file content (base64, up to 30MB), processes it in memory only, and stores nothing.
What file formats are supported? PDF and PowerPoint (.pptx) exports from Gamma.app. Legacy .ppt must be converted to .pptx first.
What happens if the watermark cannot be removed?
Some exports flatten the badge into the page image. The tool reports may_remain honestly instead of degrading your file with pixel inpainting.
Is this free? Yes — MIT-licensed open source, no account, no quota.
Web app (browser-based, no upload): gammaremover.com
CLI + Python library: gamma-watermark-remover
Local web UI: gamma-watermark-remover-webui
MCP server for Claude and AI agents: gamma-watermark-remover-mcp
Agent skill for Claude Code and OpenClaw: gamma-watermark-remover-skill
Curated Gamma resources: awesome-gamma
Only process files you created or have the right to modify. Keep originals as backup and review cleaned files before sharing. Gamma's official watermark-free route is its paid plan.
MIT
Check whether a file is a supported Gamma export format (.pdf / .pptx).
Args: file_path: path to the file to check.
Returns: Whether the file can be processed, with guidance if not.
| Name | Required | Description | Default |
|---|---|---|---|
| file_path | Yes |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It states the tool checks format and returns guidance, but does not specify behavior for missing files, permissions, or error handling. It does disclose it is a read-only check, but lacks depth.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise with a front-loaded purpose statement, followed by clearly separated Args and Returns sections. No wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool is simple with one input and an output schema provided. The description covers the main purpose and return guidance. However, it lacks details on the nature of the guidance (e.g., error codes, messages). Slight gap, but overall complete for its complexity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It adds 'path to the file to check' beyond the parameter name, but does not specify path format or existence guarantee. Adequate but not detailed.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states it checks whether a file is a supported Gamma export format (.pdf/.pptx). The verb 'check' and resource 'file' are specific. The sibling tool 'remove_gamma_watermark' suggests this is a validation step, providing differentiation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is implied as a pre-check before processing (e.g., removing watermarks), but there is no explicit guidance on when to use this tool versus alternatives like 'remove_gamma_watermark'. No when-not-to-use or prerequisites mentioned.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Remove the "Made with Gamma" watermark from a PDF or PPTX file.
Args: file_path: absolute path to the Gamma-exported .pdf or .pptx file. output_path: optional absolute path for the cleaned file. Defaults to "-no-watermark." next to the input.
Returns: A summary of what was removed and where the cleaned file was written.
| Name | Required | Description | Default |
|---|---|---|---|
| file_path | Yes | ||
| output_path | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries full burden. It discloses the removal action and output summary but omits behavioral traits like permissions, file modification side effects, or error handling. Meets minimum adequacy.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise with a clear first sentence. Args/Returns sections add structure without verbosity. Minor improvement: front-loading the purpose more prominently.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 2 parameters, no annotations, and an output schema (not shown), the description covers the core workflow. However, it lacks context on edge cases (e.g., file not found, unsupported types, large files) and prerequisites.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, yet the description adds meaningful context: file_path is absolute path, output_path is optional with default naming. This compensates well for the missing schema descriptions.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('remove') and the specific resource ('Made with Gamma' watermark from PDF/PPTX). It distinguishes from sibling tool 'check_gamma_watermark_support' by focusing on removal, not checking.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage after checking support via sibling tool but lacks explicit when-to-use, when-not-to-use, or alternative guidance. It doesn't mention prerequisites like file being Gamma-exported.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0check_gamma_watermark_supportremove_gamma_watermarkScored across 2 tools
The two tools have clearly distinct purposes: one checks file support, the other removes the watermark. No overlap in functionality.
Both tools follow a consistent verb_noun pattern with snake_case: check_gamma_watermark_support and remove_gamma_watermark.
With only 2 tools, the server is tightly scoped to its single purpose of watermark removal, and both tools are necessary for the workflow.
The tool surface covers validation (check) and action (remove), which is complete for the narrow domain of Gamma watermark removal from PDF/PPTX files.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Merge, split, extract, rotate, reorder and stamp PDF pages from your AI chat, all offline.
Download DocSend and Papermark decks as PDF or PPTX, including email-gated and protected links.
Inspect and remove C2PA content credentials from AI-generated images, video and audio. Free, no key.
High-fidelity PDF to structured Markdown conversion and document field extraction.
