Gamma

Gamma API Integration

Generate professional presentations, documents, webpages, and social posts using Gamma's AI.

Prerequisites

• A Gamma account with API access and available credits
• GAMMA_API_KEY set in your environment or Zo secrets
• Bun available to run the CLI

If you are using Zo, add GAMMA_API_KEY in Settings > Advanced.

Quick Start

# Verify API key is set
bun run Skills/gamma/scripts/gamma.ts themes

# Generate a presentation
bun run Skills/gamma/scripts/gamma.ts generate "Q4 2024 Sales Report" --mode generate --format presentation --wait

# Generate a document
bun run Skills/gamma/scripts/gamma.ts generate "AI in Healthcare" --mode generate --format document --wait

Commands

generate — Create from scratch

Generate a new gamma from AI using text input.

bun run Skills/gamma/scripts/gamma.ts generate "<inputText>" --mode <mode> [options]

Required:

• <inputText> — Content to generate from (supports text and image URLs, max 100k tokens)
• --mode <mode> — How to handle input text:
  • generate — AI creates new content based on your input (most common)
  • condense — AI summarizes/shortens your input
  • preserve — Keep your input text exactly as-is

Content Options:

• --format <type> — Output type: presentation | document | webpage | social
• --cards <n> — Number of cards/slides (1-60 Pro, 1-75 Ultra)
• --card-split <mode> — Content splitting: auto | inputTextBreaks
• --instructions <text> — Additional instructions (max 2000 chars)

Text Options:

• --language <code> — Output language (default: en)
• --tone <description> — Writing tone (e.g., "professional", "casual", "academic")
• --audience <description> — Target audience (e.g., "executives", "students", "developers")
• --amount <level> — Detail level: brief | medium | detailed | extensive

Image Options:

• --images <source> — Image source: aiGenerated | webSearch | none
• --image-model <model> — AI model for image generation (see models below)
• --image-style <description> — Visual style (e.g., "minimalist", "vibrant", "corporate")

Layout Options:

• --dimensions <size> — Card dimensions: fluid | 16x9 | 4x3 | 1x1 | 4x5 | 9x16 | 3x4 | 2x3 | 5x4 | 21x9

Organization:

• --theme <id> — Theme ID (use themes command to list available)
• --folders <id1,id2> — Folder IDs for storage (comma-separated)
• --export <format> — Also export as: pdf | pptx

Sharing:

• --visibility <level> — Access level: private | public | unlisted
• --allow-copy — Allow viewers to copy content
• --allow-duplication — Allow viewers to duplicate the gamma

Behavior:

• --wait — Poll until generation completes, then return final URLs
• --timeout <seconds> — Max wait time (default: 300)

from-template — Create from existing gamma

Create a new gamma by adapting an existing template.

bun run Skills/gamma/scripts/gamma.ts from-template <gammaId> "<prompt>" [options]

Required:

• <gammaId> — Template gamma ID (find in Gamma app: ⋮ > Copy gammaId for API)
• <prompt> — Instructions for adapting the template (max 100k tokens)

Available Options: Same as generate except these are NOT available: --mode, --format, --cards, --card-split, --language, --tone, --audience, --amount, --dimensions

status — Check generation status

bun run Skills/gamma/scripts/gamma.ts status <generationId>

Returns status (pending | completed | failed), gamma URL, and export URLs if ready.

themes — List available themes

bun run Skills/gamma/scripts/gamma.ts themes

Returns standard themes and your custom themes with their IDs.

folders — List your folders

bun run Skills/gamma/scripts/gamma.ts folders

Returns your Gamma folders with their IDs.

Image Models

Budget (2 credits/image)

Standard (8-15 credits/image)

Premium (20-33 credits/image)

Ultra-only (30-120 credits/image)

Languages

English Variants

• en — English (US) — default
• en-gb — English (UK)
• en-in — English (India)

Spanish Variants

• es — Spanish (general)
• es-es — Spanish (Spain)
• es-mx — Spanish (Mexico)
• es-419 — Spanish (Latin America)

Chinese

• zh-cn — Simplified Chinese
• zh-tw — Traditional Chinese

Japanese

• ja — Japanese (です/ます polite style)
• ja-da — Japanese (だ/である plain style)

Other Major Languages

fr (French), de (German), it (Italian), pt-br (Portuguese Brazil), pt-pt (Portuguese Portugal), ko (Korean), ar (Arabic), hi (Hindi), ru (Russian), nl (Dutch), pl (Polish), tr (Turkish), vi (Vietnamese), th (Thai), id (Indonesian)

Full List

60+ languages supported including: af, sq, ar-sa, bn, bs, bg, ca, hr, cs, da, et, fi, el, gu, ha, he, hu, is, kn, kk, lv, lt, mk, ms, ml, mr, nb, fa, ro, sr, sl, sw, sv, tl, ta, te, uk, ur, uz, cy, yo

Examples

Basic Presentation

bun run Skills/gamma/scripts/gamma.ts generate "Quarterly sales report for Q4 2024. Include revenue growth, key wins, and 2025 outlook." \
  --mode generate \
  --format presentation \
  --wait

Detailed Document

bun run Skills/gamma/scripts/gamma.ts generate "The impact of artificial intelligence on healthcare delivery" \
  --mode generate \
  --format document \
  --tone "academic" \
  --audience "medical professionals" \
  --amount detailed \
  --images aiGenerated \
  --image-model imagen-4-pro \
  --wait

Startup Pitch Deck

bun run Skills/gamma/scripts/gamma.ts generate "TechStartup is revolutionizing supply chain with AI. Founded 2023, $2M ARR, seeking Series A." \
  --mode generate \
  --format presentation \
  --cards 12 \
  --tone "confident and compelling" \
  --audience "venture capitalists" \
  --images aiGenerated \
  --image-style "modern tech aesthetic" \
  --wait

Webpage

bun run Skills/gamma/scripts/gamma.ts generate "Our company story: from garage startup to industry leader" \
  --mode generate \
  --format webpage \
  --images aiGenerated \
  --image-model flux-1-pro \
  --image-style "professional corporate photography" \
  --visibility public \
  --wait

Social Post

bun run Skills/gamma/scripts/gamma.ts generate "Announcing our new product launch! Revolutionary features that will change how teams collaborate." \
  --mode generate \
  --format social \
  --tone "exciting and engaging" \
  --wait

From Existing Content (Preserve Text)

# When you have exact content you want to use
bun run Skills/gamma/scripts/gamma.ts generate "$(cat my-prepared-content.md)" \
  --mode preserve \
  --format presentation \
  --wait

Condense Long Content

# Summarize a long document into a short presentation
bun run Skills/gamma/scripts/gamma.ts generate "$(cat long-report.txt)" \
  --mode condense \
  --format presentation \
  --cards 5 \
  --instructions "Focus on key findings and recommendations only" \
  --wait

Adapt a Template

# Use an existing gamma as a template
bun run Skills/gamma/scripts/gamma.ts from-template g_xo2yqze5mj8z62g \
  "Adapt this pitch deck for a healthcare AI startup instead of fintech" \
  --images aiGenerated \
  --wait

Non-English Output

# Japanese presentation
bun run Skills/gamma/scripts/gamma.ts generate "Company introduction and services overview" \
  --mode generate \
  --format presentation \
  --language ja \
  --wait

# Spanish document
bun run Skills/gamma/scripts/gamma.ts generate "Quarterly financial report" \
  --mode generate \
  --format document \
  --language es-mx \
  --wait

Export to PDF/PPTX

bun run Skills/gamma/scripts/gamma.ts generate "Team training materials" \
  --mode generate \
  --format presentation \
  --export pdf \
  --wait

Tips

1. Text Mode Selection:

   • Use generate for most cases — AI creates content from your brief
   • Use preserve when you have exact content already written
   • Use condense to summarize long documents

2. Image Quality vs. Cost:

   • Start with budget models (flux-1-quick, imagen-3-flash) for drafts
   • Use premium models for final versions

3. Finding Template IDs:

   • In Gamma app: Open template → ⋮ menu → "Copy gammaId for API"

4. Finding Theme IDs:

   • Run bun run Skills/gamma/scripts/gamma.ts themes
   • Or in Gamma app: Themes → ⋮ menu → "Copy themeId for API"

5. Polling Behavior:

   • Without --wait: Returns immediately with generationId
   • With --wait: Polls every 5 seconds until complete (default timeout: 5 minutes)

References

• Gamma API Overview
• Generate API Parameters
• Template API Parameters
• Image Models
• Languages