Gamma Mcp Server

io.github.amer-progPublished by GitHub account amer-progv1.0.0Updated Nov 3, 2026

Create presentations, documents, and webpages from any MCP-compatible AI assistant via Gamma.app

VerifiedSTDIODesktop onlyProductivity & WorkflowMedia & Design

Overview

AI-generated overview

Connects an AI assistant to Gamma.app so it can generate presentations, documents, webpages and social posts, and manage the results.

What you can do
  • Generate presentations, documents, webpages and social posts from a prompt, with format and size options
  • Check generation progress, optionally polling until completion, and verify server and API reachability
  • Remix existing Gamma templates with new content or variable substitutions
  • Browse available visual themes and list your Gamma folders
  • Share generated content by email and archive items from your workspace
When to use it
Useful when you want an assistant to draft slide decks, documents, webpages or social posts in Gamma directly from a conversation, or to rework existing Gamma templates and share the output without leaving the chat.
Requirements
Runs locally as a Node.js 18+ process built from the npm package or repository. Needs a Gamma API key in the GAMMA_API_KEY environment variable, which requires a Gamma Pro, Ultra, Team or Business account. Optional variables override the API base URL and polling interval and attempt limits.
Before you install
The Gamma API key is a secret placed in the assistant's MCP config file; keep that file private. Generations consume Gamma API credits, so usage costs money and can run out. The server creates content, sends email shares and archives items in your Gamma workspace, which changes data. Prompts and content are sent to Gamma's API.

Installation

In SourceWeft

  1. Open Gamma Mcp Server in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Gamma MCP Server

[License: MIT] [Node.js Version] [TypeScript] [npm version] [npm downloads] [CI] [MCP SDK]

A Model Context Protocol (MCP) server that integrates Gamma.app with AI assistants. Create presentations, documents, webpages, and social posts directly from your AI conversations.

Works with: Claude Code, Claude Desktop, OpenCode, GitHub Copilot CLI, Google Gemini CLI, and any other MCP-compatible AI assistant.

Features

  • Generate Content: Create professional presentations, documents, webpages, and social posts from text prompts
  • Theme Support: Browse and apply visual themes to your content
  • Folder Organization: Save generated content to specific folders
  • Template Remix: Create variations of existing Gamma templates
  • Email Sharing: Share generated content directly via email

Quick Start

1. Clone and Install

bash
git clone https://github.com/Arkava-AI/gamma-mcp-server.gitcd gamma-mcp-servernpm installnpm run build

2. Get Your Gamma API Key

  1. Log in to gamma.app
  2. Go to Settings > API (or Settings > Members > API tab)
  3. Click Create API key
  4. Copy the key (format: sk-gamma-xxxxxxxx)

Note: Requires Gamma Pro, Ultra, Team, or Business account.

3. Configure Your AI Assistant

Choose your AI assistant below for setup instructions.


Claude Desktop (macOS / Windows / Linux)

Config File

OSPath
macOS~/Library/Application Support/Claude/claude_desktop_config.json
Linux~/.config/Claude/claude_desktop_config.json
Windows%APPDATA%\Claude\claude_desktop_config.json

Add to the mcpServers object:

json
{  "mcpServers": {    "gamma": {      "command": "node",      "args": ["/absolute/path/to/gamma-mcp-server/dist/index.js"],      "env": {        "GAMMA_API_KEY": "sk-gamma-your-api-key-here"      }    }  }}

Restart Claude Desktop

Restart Claude Desktop to load the new MCP server. You should see "gamma" in your MCP servers list.


Claude Code

Claude Code uses the same MCP configuration as Claude Desktop. If you've already configured Claude Desktop, you're all set.

For project-level configuration, create a .claude/settings.json file in your project directory with the same mcpServers structure shown above. This allows different projects to use different MCP server configurations.


OpenCode

Config File

ScopePath
Global (user)~/.config/opencode/opencode.json
Project./opencode.json (in your project root)

JSON Structure

json
{  "mcp": {    "gamma": {      "type": "local",      "command": ["node", "/absolute/path/to/gamma-mcp-server/dist/index.js"],      "enabled": true,      "environment": {        "GAMMA_API_KEY": "sk-gamma-your-api-key-here"      }    }  }}

Note: OpenCode uses a different config format from Claude Desktop — mcp (not mcpServers), type field required ("local" or "remote"), command is an array, and env vars go under environment.

Restart OpenCode after editing the config to load the server.


GitHub Copilot CLI

Config File

  • ~/.copilot/mcp-config.json

JSON Structure

json
{  "mcpServers": {    "gamma": {      "type": "local",      "command": "node",      "args": ["/absolute/path/to/gamma-mcp-server/dist/index.js"],      "env": {        "GAMMA_API_KEY": "sk-gamma-your-api-key-here"      },      "tools": ["*"]    }  }}

Note: Requires the GitHub Copilot CLI (gh copilot) — not the same as OpenAI Codex.


OpenAI Codex

Config File

  • ~/.codex/config.toml (TOML format, not JSON)

TOML Structure

toml
[mcp_servers.gamma]command = "node"args = ["/absolute/path/to/gamma-mcp-server/dist/index.js"]enabled = true
[mcp_servers.gamma.env]GAMMA_API_KEY = "sk-gamma-your-api-key-here"

Note: Codex uses TOML format, not JSON. The env section is a separate table under [mcp_servers.gamma.env].


Google Gemini CLI

Config File

ScopePath
User~/.gemini/settings.json
Project.gemini/settings.json (in your project root)

JSON Structure

json
{  "mcpServers": {    "gamma": {      "command": "node",      "args": ["/absolute/path/to/gamma-mcp-server/dist/index.js"],      "cwd": "/absolute/path/to/gamma-mcp-server",      "env": {        "GAMMA_API_KEY": "sk-gamma-your-api-key-here"      },      "timeout": 30000    }  }}

Restart Gemini CLI after editing the config to load the server.


Available Tools

ToolDescription
gamma_generateCreate presentations, documents, webpages, or social posts
gamma_get_statusCheck generation progress (with optional polling)
gamma_from_templateRemix existing Gamma templates
gamma_list_themesBrowse available visual themes
gamma_list_foldersList your Gamma folders
gamma_share_emailShare content via email
gamma_healthVerify server and API are reachable
gamma_archiveArchive a Gamma from your workspace

gamma_generate

Create new content using Gamma's AI.

Formats & Sizes:

  • presentation: fluid, 16x9, 4x3
  • document: fluid, pageless, letter, a4
  • social: 1x1, 4x5, 9x16
  • webpage: fluid

Example prompts in your AI assistant:

  • "Create a 5-slide presentation about sustainable energy"
  • "Generate a document explaining our Q1 results"
  • "Make a social media post announcing our new product"

gamma_get_status

Check if a generation has completed. Set waitForCompletion: true to automatically poll until done.

gamma_from_template

Remix an existing Gamma with new content or variable substitutions.

json
{  "templateId": "gamma_xyz789",  "prompt": "Update for Q1 2025",  "variables": { "company_name": "Acme Corp" }}

Multi-Machine Setup

This repository is designed for easy deployment across multiple machines:

bash
# On each machine:git clone https://github.com/Arkava-AI/gamma-mcp-server.gitcd gamma-mcp-servernpm install && npm run build
# Then configure your AI assistant with the local path

To update on any machine:

bash
git pullnpm installnpm run build# Restart your AI assistant

Environment Variables

VariableDefaultDescription
GAMMA_API_KEY(required)Your Gamma API key (sk-gamma-...)
GAMMA_API_BASE_URLhttps://public-api.gamma.app/v1.0Override for self-hosted Gamma instances
GAMMA_POLL_INTERVAL_MS2000Milliseconds between status polls (default 2s)
GAMMA_MAX_POLL_ATTEMPTS150Max polling attempts before timeout (default 150 × 2s = 5 min)

Development

bash
# Run in development mode with auto-reloadnpm run dev
# Build for productionnpm run build
# Run lintingnpm run lint
# Format codenpm run format
# Type checknpm run typecheck
# Test with MCP Inspectornpm run inspect

Project Structure

gamma-mcp-server/├── src/│   ├── index.ts          # Main entry point│   ├── constants.ts      # Configuration constants│   ├── types.ts          # TypeScript interfaces│   ├── schemas/          # Zod validation schemas│   ├── services/         # API client and formatters│   └── tools/            # MCP tool implementations├── dist/                 # Compiled JavaScript (generated)├── package.json├── tsconfig.json└── eslint.config.js

API Credits

Gamma uses a credit-based system for API usage. Credits are consumed per generation. Monitor your usage in the Gamma dashboard and enable auto-recharge if needed.


Troubleshooting

ErrorSolution
"GAMMA_API_KEY environment variable is required"Ensure env.GAMMA_API_KEY is set in your AI assistant's MCP config
"Invalid API key"Keys should start with sk-gamma-. Verify the complete key.
"Rate limit exceeded"Wait a few minutes. Contact Gamma support for higher limits.
"Insufficient credits"Top up credits or enable auto-recharge in Gamma settings.

Requirements

  • Node.js 18+
  • Gamma Pro/Ultra/Team/Business account (for API access)
  • MCP-compatible AI assistant (Claude Code, Claude Desktop, OpenCode, GitHub Copilot CLI, Gemini CLI, etc.)

Maintainer

Arkava Ltd — [email protected]


License

MIT License - see LICENSE for details.


Links

Source: README.md at commit 6e93f04

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.0.0LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts