gamma-mcp
Enables AI-powered generation of presentations, documents, social posts, and webpages via Gamma.app.
README
Gamma MCP Server
MCP server for Gamma.app - AI-powered presentation, document, and website generation.
Features
- Generate from scratch - Create presentations, documents, social posts, or webpages from text
- Create from template - Adapt existing Gamma templates with new content
- List themes - Browse available themes for styling
- List folders - Browse workspace folders for organization
- Poll generation status - Check when async generation completes
Installation
npm install @purple-horizons/gamma-mcp
Or clone and build:
git clone https://github.com/Purple-Horizons/gamma-mcp.git
cd gamma-mcp
npm install
npm run build
Configuration
Environment Variable
Set your Gamma API key:
export GAMMA_API_KEY=sk-gamma-xxxxxxxx
Get your API key from Gamma Account Settings (requires Pro, Ultra, Teams, or Business plan).
Claude Desktop / OpenClaw
Add to your MCP config:
{
"mcpServers": {
"gamma": {
"command": "node",
"args": ["/path/to/gamma-mcp/dist/index.js"],
"env": {
"GAMMA_API_KEY": "sk-gamma-xxxxxxxx"
}
}
}
}
mcporter
Add to ~/.config/mcporter/mcporter.json:
{
"servers": {
"gamma": {
"command": "node",
"args": ["/path/to/gamma-mcp/dist/index.js"],
"env": {
"GAMMA_API_KEY": "sk-gamma-xxxxxxxx"
}
}
}
}
Tools
gamma_generate
Generate a new gamma from text input.
Parameters:
inputText(required) - Content to generate from (text + image URLs)textMode(required) -generate|condense|preserveformat-presentation|document|social|webpagethemeId- Theme ID (fromgamma_list_themes)numCards- Number of slides (1-60 for Pro, 1-75 for Ultra)cardSplit-auto|inputTextBreaksadditionalInstructions- Extra generation guidancefolderIds- Array of folder IDs to save toexportAs-pdf|pptxtextAmount-brief|medium|detailed|extensivetextTone- Tone description (e.g., "professional, inspiring")textAudience- Target audiencetextLanguage- Language code (default: "en")imageSource-aiGenerated|pictographic|pexels|giphy|webAllImages|webFreeToUse|webFreeToUseCommercially|placeholder|noImagesimageModel- AI model for images (e.g., "flux-1-pro", "imagen-4-pro")imageStyle- Image style descriptiondimensions- Card dimensions (format-dependent)
Returns: { generationId: "..." }
gamma_get_generation
Check generation status.
Parameters:
generationId(required) - ID from gamma_generate
Returns:
{
"status": "pending" | "completed",
"generationId": "...",
"gammaUrl": "https://gamma.app/docs/...",
"credits": { "deducted": 150, "remaining": 3000 }
}
gamma_create_from_template
Create a gamma based on an existing template.
Parameters:
gammaId(required) - Template gamma IDprompt(required) - Instructions and contentthemeId- Override themefolderIds- Destination foldersexportAs-pdf|pptximageModel- AI model for new imagesimageStyle- Style for new images
Returns: { generationId: "..." }
gamma_list_themes
List available themes.
Parameters:
query- Search by namelimit- Max results (default 50)after- Pagination cursor
Returns:
{
"data": [{ "id": "...", "name": "...", "type": "standard|custom" }],
"hasMore": false,
"nextCursor": null
}
gamma_list_folders
List workspace folders.
Parameters:
query- Search by namelimit- Max results (default 50)after- Pagination cursor
Returns:
{
"data": [{ "id": "...", "name": "..." }],
"hasMore": false,
"nextCursor": null
}
Usage Example
// 1. Generate a presentation
const gen = await gamma_generate({
inputText: "5 ways AI is transforming healthcare",
textMode: "generate",
format: "presentation",
numCards: 10,
imageSource: "aiGenerated",
imageStyle: "modern, professional photography"
});
// 2. Poll for completion
let status;
do {
await sleep(5000);
status = await gamma_get_generation({ generationId: gen.generationId });
} while (status.status === "pending");
// 3. Use the result
console.log(status.gammaUrl); // https://gamma.app/docs/xxxxx
Credits
Gamma uses a credit-based system:
- ~3-4 credits per card
- ~2-120 credits per image (depends on model)
Check your remaining credits in the generation response.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。