How to Connect Gamma to Claude via MCP

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → Gamma, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

Gamma + Claude

Let your AI create Gamma presentations, documents, webpages and social posts from text or from a template, follow the generation until it is ready, export to PDF, PPTX or PNG, and find themes, folders, templates and existing gammas.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

GAMMA_API_KEY

Step 1 — Get credentials

1. Sign in at https://gamma.app. API keys exist only on the paid Pro, Ultra, Teams and Business plans.
2. Open Settings → API Keys (https://gamma.app/settings/api-keys) and create a key. It starts with sk-gamma-.
3. Paste it into GAMMA_API_KEY and install the connector. The key is sent only to Gamma.

Billed to your Gamma credits: every new gamma spends credits from your Gamma workspace. Gamma's current figures, which can change:

• 1 to 3 credits per card, slightly more per card when you start from a template.
• 2 to 125 credits per AI image, depending on the image model.
• A 10-card deck with 5 basic images costs about 20 to 60 credits.

For a cheap draft, ask for few cards and no AI images (no images, Pexels stock photos or placeholders). When a generation is done, the AI tells you the credits it used and what is left.

Confirm first: before gamma_generate, gamma_generate_multi_page or gamma_generate_from_template, the AI checks the topic, format, number of cards and image source with you. It also checks the gamma and the file format before an export.

How a generation works: Gamma builds the gamma in the background. The AI starts it, gets an id right away and checks gamma_get_generation every 5 to 10 seconds until it is completed or failed. A small deck takes under a minute; 40 cards with AI images can take several minutes. The result has the link to the gamma, the credits used and, if you asked for an export, a download link. Existing gammas are exported the same way: gamma_export_gamma starts the export and gamma_get_export returns the file once it is ready.

What it can make:

• Presentations (16:9, 4:3 or fluid), documents (pageless, Letter or A4), webpages and social posts (1:1, 4:5 or 9:16), with 1 to 75 cards (10 by default).
• From your text, Gamma can expand a short brief, condense a long text or keep your words as they are.
• One file with up to 50 pages under a single link, optionally published as a Gamma site.
• A new gamma built on one of your workspace's one-page templates, rewritten for your prompt. gamma_search_templates also shows Gamma's official templates, but those are for inspiration only.
• Exports to PDF, PPTX or PNG (a .zip with one image per card).

Look-ups: themes and folders by name, so the AI can use your brand theme and save into the right folder; existing gammas by words in their title and text, or by who made them and when; and one gamma's details (title, link, theme, author, dates), not its slide content.

Download links: an export link works for anyone who has it for about a week, without your key. Share it only with the people who should see the file.

Time limit: a call through the connector stops after 30 seconds, so the AI always checks back instead of waiting.

Common errors: 401: the key is wrong or deleted, or the plan has no API access. 402: the workspace is out of credits; buy credits or turn on auto-recharge in Gamma's billing settings. 403: the plan does not include the feature, or your user may not open that gamma or folder. 400: an invalid option, for example a theme id that does not exist; the message names the field. 429: too many requests; the AI waits about 30 seconds and retries once. 404 on a status check: the id is unknown or belongs to another workspace.

The connector does not use Gamma's agent, editing of existing gammas, standalone image generation, comments, analytics, or archiving and deleting gammas.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in Claude

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

Recommended path: works on claude.ai web without editing any config file.

1. In Claude, open Customize → Connectors.
2. Click +, then Add custom connector.
3. Enter a name, e.g. Gamma, and the remote MCP server URL https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID, then click Add.
4. Click Connect and sign in to AnythingMCP (OAuth) to allow access.

All tools below appear in your chat: start typing prompts.

Available tools

Example prompts

• "Make a 10-card presentation from these notes about our Q4 roadmap, in our brand theme and without AI images, and tell me how many credits it used."
• "Turn this blog post into a 9:16 social post with Pexels photos and give me the link."
• "Find the gammas I updated this month and export the sales kickoff deck to PPTX."
• "Which templates do we have in our workspace? Use the case study template to make a version for our customer Nordlicht Logistik."

FAQ

Does it work with Claude Code as well as Claude Desktop? Yes, point both at https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID.

Next steps

• Connector landing page
• Connect to ChatGPT
• Connect to Cursor
• Connect to Gemini
• Connect to Meta Muse