Gamma Presentations Skill

Generate professional presentations, documents, and websites using the Gamma API.

Gamma is an AI-powered platform with 50M+ users for creating presentations, documents, social posts, and websites. This skill enables Alex to generate polished content programmatically.

When to Use

User asks to create a presentation or slide deck
Need to generate a document or report
Creating social media content (carousels, stories)
Building simple webpages from content
Converting notes or outlines into polished presentations

Prerequisites

Gamma account (Pro, Ultra, Teams, or Business plan for API access)
API key from gamma.app/settings
Environment variable: GAMMA_API_KEY

Help & User Manual

How to Ask Alex to Create Gamma Content

Simply tell Alex what you want to create. Alex will write the actual content for you, then generate the presentation.

From a Topic (Alex writes all content)

"Create a 10-slide presentation about machine learning for executives"
"Make a pitch deck for my startup idea: [description]"
"Generate a social media carousel about productivity tips"

What Alex does:

Creates a detailed markdown file with real slide content
Sends it to Gamma API for professional formatting
Returns the link and/or downloads the file
Opens it automatically if requested

From Your Own Content File

"Create a presentation from README.md"
"Turn ROADMAP-UNIFIED.md into a 15-slide deck"
"Make a document from alex_docs/COGNITIVE-ARCHITECTURE.md"

From Reference Documents

"Create a presentation about appropriate reliance using all the documents in the article folder"
"Make a pitch deck from my business plan - use investor-deck.md and financials.md"

Alex will:

Read and synthesize the referenced files
Write presentation content based on that material
Generate the formatted presentation

With Specific Options

"Create a presentation from my-notes.md with 12 slides, professional tone, for investors"
"Generate a webpage from PROJECT.md in Portuguese"
"Make an Instagram carousel (4x5) from tips.txt with vibrant AI images"

Command Quick Reference

What You Say	What Happens
"Create a presentation about X"	Alex writes content, then generates slides
"Create a presentation from FILE"	Reads file, converts to slides
"Create presentation using [files]"	Alex synthesizes files into slides
"Make a document from FILE"	Creates paginated document
"Generate a social post from X"	Creates carousel/story format
"Create a webpage from FILE"	Generates simple website

Available Options

When asking Alex to create Gamma content, you can specify:

Option	Values	Example
Format	presentation, document, social, webpage	"as a document"
Slides/Cards	1-75	"with 12 slides"
Tone	any description	"professional and confident"
Audience	any description	"for investors"
Language	60+ languages	"in Spanish"
Dimensions	16x9, 4x3, 1x1, 4x5, 9x16, letter, a4	"in 16x9 format"
Images	AI, Unsplash, no images	"with AI-generated images"
Export	pptx, pdf	"export as PowerPoint"
Illustrations only		"use only illustrations" (no photos)

Step-by-Step: Create from a Workspace File

Tell Alex the file path:

"Create a presentation from alex_docs/USER-MANUAL.md"
Alex reads the file and sends content to Gamma API
Alex polls for completion (usually 15-60 seconds)
Alex returns the link: https://gamma.app/docs/xxxxx
Open the link to view, edit, or download your presentation

Customization Examples

Basic:

"Create a presentation from README.md"

With slide count:

"Create a 15-slide presentation from README.md"

With audience:

"Create a presentation from README.md for new developers, friendly tone"

With export:

"Create a presentation from ROADMAP.md and export as PowerPoint"

Full customization:

"Create a 12-slide pitch deck from my-startup.md in 16x9 format, confident tone, for VCs, with modern AI images, export as PDF and PowerPoint"

What Alex Does Behind the Scenes

Reads the specified file from your workspace
Sends to Gamma API with your options:
- textMode: "generate" (expands brief content) or "condense" (summarizes long content)
- format, numCards, textOptions, imageOptions based on your request
Polls generation status every 2-3 seconds
Returns the Gamma URL when complete
Downloads export (PPTX/PDF) if requested
Auto-opens the file if --open flag is used
Reports credits used and remaining balance

Pro Tips for Great Results

1. Structure Your Content with Slide Markers

Use clear headers in your markdown — they become natural slide breaks:

## Slide 1: Title
Main message here

## Slide 2: The Problem
- Pain point 1
- Pain point 2

## Slide 3: Our Solution
Description of solution

2. Use Illustrations for Professional Decks

Specify --image-style with "illustrations" for consistent, modern look:

--image-model ideogram --image-style "modern clean illustrations, abstract concepts"

3. Match Tone to Audience

Audience	Suggested Tone
Executives	"professional and concise"
Researchers	"intellectual and evidence-based"
Developers	"technical and practical"
Investors	"confident and visionary"
General	"friendly and accessible"

4. One Command, Open Result

Always use --open to immediately review your deck:

node scripts/gamma-generator.js -f content.md -e pptx --open

5. Iterate with Cost-Effective Models

Use cheaper models while iterating, premium for final:

# Drafts (2 credits/image)
--image-model flux-quick

# Final version (20 credits/image)
--image-model ideogram

Cost Awareness

Alex will inform you of credit usage after each generation:

Complexity	Estimated Credits
Simple 5-card deck	~20-25 credits
Standard 10-card presentation	~40-50 credits
20-card with premium images	~200-400 credits
Full pitch deck with exports	~100-200 credits

Check your balance: gamma.app/settings/billing

Quick Reference

API Base URL

https://public-api.gamma.app

Authentication

--header 'X-API-KEY: <your-api-key>'
--header 'Content-Type: application/json'

Core Endpoints

Endpoint	Method	Description
`/v0.2/generations`	POST	Generate new gamma from text
`/v0.2/generations/{id}`	GET	Check status, get URLs
`/v0.2/themes`	GET	List available themes

Generate API Parameters

Required Parameters

Parameter	Type	Description
`inputText`	string	Content to generate from (max 100k tokens / ~400k chars)
`textMode`	enum	`generate` (expand), `condense` (summarize), `preserve` (keep exact)

Format Options

Parameter	Default	Values
`format`	`presentation`	`presentation`, `document`, `social`, `webpage`
`numCards`	10	1-60 (Pro) or 1-75 (Ultra)
`cardSplit`	`auto`	`auto`, `inputTextBreaks` (use `\n---\n`)

Text Options

"textOptions": {
  "amount": "medium",      // brief, medium, detailed, extensive
  "tone": "professional",  // free text, 1-500 chars
  "audience": "executives", // free text, 1-500 chars
  "language": "en"         // 60+ languages
}

Image Options

"imageOptions": {
  "source": "aiGenerated",  // see source options below
  "model": "flux-1-pro",    // see model list below
  "style": "modern, minimal" // free text, 1-500 chars
}

Image Sources:

aiGenerated - AI-generated images (use with model + style)
pictographic - Pictographic library
unsplash - Unsplash photos
giphy - Animated GIFs
webAllImages - Web images (any license)
webFreeToUse - Personal use licensed
webFreeToUseCommercially - Commercial use licensed
placeholder - Empty placeholders
noImages - No images (use if providing URLs in inputText)

Card Dimensions

Format	Options
Presentation	`fluid`, `16x9`, `4x3`
Document	`fluid`, `pageless`, `letter`, `a4`
Social	`1x1`, `4x5` (Instagram/LinkedIn), `9x16` (Stories)

Export Options

"exportAs": "pptx"  // or "pdf"

AI Image Models

Cost-Effective (2 credits/image)

Model	API Value
Flux Fast 1.1	`flux-1-quick`
Flux Kontext Fast	`flux-kontext-fast`
Imagen 3 Fast	`imagen-3-flash`
Luma Photon Flash	`luma-photon-flash-1`

Standard (8-15 credits/image)

Model	API Value	Credits
Flux Pro	`flux-1-pro`	8
Imagen 3	`imagen-3-pro`	8
Ideogram 3 Turbo	`ideogram-v3-turbo`	10
Leonardo Phoenix	`leonardo-phoenix`	15

Premium (20-33 credits/image)

Model	API Value	Credits
Ideogram 3	`ideogram-v3`	20
Imagen 4	`imagen-4-pro`	20
Gemini 2.5 Flash	`gemini-2.5-flash-image`	20
Recraft	`recraft-v3`	20
GPT Image	`gpt-image-1-medium`	30
DALL-E 3	`dall-e-3`	33

Ultra (30-120 credits/image, Ultra plan only)

Model	API Value	Credits
Flux Ultra	`flux-1-ultra`	30
Imagen 4 Ultra	`imagen-4-ultra`	30
Recraft Vector	`recraft-v3-svg`	40
GPT Image Detailed	`gpt-image-1-high`	120

Credit Costs

Factor	Credits
Per card	3-4
Basic AI image	~2
Standard AI image	~8-15
Premium AI image	~20-40
Ultra AI image	~40-120

Estimates:

10-card deck + 5 basic images = ~40-50 credits
20-card doc + 15 premium images = ~360-680 credits

Example Requests

Simple Presentation

curl -X POST https://public-api.gamma.app/v1.0/generations \
  -H 'Content-Type: application/json' \
  -H 'X-API-KEY: sk-gamma-xxx' \
  -d '{
    "inputText": "Introduction to machine learning for business leaders",
    "textMode": "generate",
    "format": "presentation",
    "numCards": 8
  }'

Professional Pitch Deck

curl -X POST https://public-api.gamma.app/v1.0/generations \
  -H 'Content-Type: application/json' \
  -H 'X-API-KEY: sk-gamma-xxx' \
  -d '{
    "inputText": "Our startup solves remote team collaboration with AI-powered async video messaging. Founded 2024, 10k users, $500k ARR, seeking $2M seed round.",
    "textMode": "generate",
    "format": "presentation",
    "numCards": 12,
    "textOptions": {
      "amount": "medium",
      "tone": "confident, professional, visionary",
      "audience": "venture capital investors"
    },
    "imageOptions": {
      "source": "aiGenerated",
      "model": "flux-1-pro",
      "style": "modern tech, clean, professional photography"
    },
    "cardOptions": {
      "dimensions": "16x9"
    },
    "exportAs": "pptx"
  }'

Document from Notes

curl -X POST https://public-api.gamma.app/v1.0/generations \
  -H 'Content-Type: application/json' \
  -H 'X-API-KEY: sk-gamma-xxx' \
  -d '{
    "inputText": "Meeting notes from Q4 planning...\n---\nBudget discussion...\n---\nAction items...",
    "textMode": "preserve",
    "format": "document",
    "cardSplit": "inputTextBreaks",
    "textOptions": {
      "language": "en"
    },
    "imageOptions": {
      "source": "noImages"
    }
  }'

Social Media Carousel

curl -X POST https://public-api.gamma.app/v1.0/generations \
  -H 'Content-Type: application/json' \
  -H 'X-API-KEY: sk-gamma-xxx' \
  -d '{
    "inputText": "5 productivity tips for developers: 1. Time blocking 2. Pomodoro technique 3. Code reviews in batches 4. Automate repetitive tasks 5. Take real breaks",
    "textMode": "generate",
    "format": "social",
    "numCards": 6,
    "cardOptions": {
      "dimensions": "4x5"
    },
    "imageOptions": {
      "source": "aiGenerated",
      "model": "ideogram-v3-turbo",
      "style": "vibrant, modern, tech aesthetic"
    }
  }'

Response Handling

POST Response

{
  "generationId": "gen_abc123",
  "status": "pending"
}

GET Status Response (Pending)

{
  "generationId": "gen_abc123",
  "status": "pending"
}

GET Status Response (Completed)

{
  "generationId": "gen_abc123",
  "status": "completed",
  "gammaUrl": "https://gamma.app/docs/xyz",
  "pptxUrl": "https://...",  // if exportAs: "pptx"
  "pdfUrl": "https://...",   // if exportAs: "pdf"
  "creditsUsed": 45
}

Polling Pattern

async function waitForGeneration(generationId, apiKey) {
  const maxAttempts = 30;
  const delayMs = 2000;

  for (let i = 0; i < maxAttempts; i++) {
    const response = await fetch(
      `https://public-api.gamma.app/v1.0/generations/${generationId}`,
      { headers: { 'X-API-KEY': apiKey } }
    );
    const data = await response.json();

    if (data.status === 'completed') return data;
    if (data.status === 'failed') throw new Error(data.error);

    await new Promise(r => setTimeout(r, delayMs));
  }
  throw new Error('Generation timeout');
}

MCP Integration

Gamma provides a hosted MCP server for AI tool integration.

Setup in Claude

Open Claude Desktop or Web
Settings → Connectors
Search "Gamma" → Connect
Authorize Gamma account access

MCP Tools Available

Tool	Capability
`generate_content`	Create presentations, docs, webpages, social
`browse_themes`	Search theme library
`organize_folders`	Save to workspace folders

Effective MCP Prompts

Good:

"Create a 10-slide marketing strategy presentation covering target audience, campaign channels, budget breakdown, and success metrics. Use a professional blue theme and modern photography style."

Better:

"Create a pitch deck for investors about our AI startup. 12 slides, 16:9 format, professional tone. Include: problem, solution, market size, business model, traction, team, financials, ask. Export as PowerPoint."

Troubleshooting

Insufficient Credits

Problem: Error "insufficient credits"

Solution:

Check balance at gamma.app/settings/billing
Purchase credits or upgrade plan
Enable auto-recharge

Authentication Failed

Problem: 401 Unauthorized

Solution:

Verify API key is correct
Check plan supports API (Pro+)
Regenerate key if compromised

Generation Timeout

Problem: Status stays "pending" too long

Solution:

Complex generations take longer (large numCards, premium images)
Increase polling timeout to 2-3 minutes
Simplify request (fewer cards, basic images)

Token Limit Exceeded

Problem: Input too large

Solution:

inputText max: 100k tokens (~400k chars)
Split into multiple generations
Use textMode: "condense" for long content

Best Practices

Start simple - Use defaults, then customize
Match model to need - Basic images for drafts, premium for finals
Use appropriate textMode:
- generate: Brief input → expanded content
- condense: Long input → summarized content
- preserve: Keep exact text
Control costs - Use flux-1-quick or imagen-3-flash for iteration
Export wisely - Only request PPTX/PDF when needed (adds processing time)
Organize content - Use folderIds for workspace organization

CLI Script

A standalone Node.js script is available for command-line generation:

Location: scripts/gamma-generator.js

Quick Examples

# Simple topic
node scripts/gamma-generator.js --topic "Introduction to AI"

# From file with PowerPoint export
node scripts/gamma-generator.js --file README.md --export pptx

# Generate and immediately open in PowerPoint 🚀
node scripts/gamma-generator.js --file README.md --export pptx --open

# Full customization with auto-open
node scripts/gamma-generator.js \
  --file my-content.md \
  --slides 15 \
  --tone "professional and inspiring" \
  --audience "executives" \
  --image-model ideogram \
  --image-style "modern illustrations, clean" \
  --dimensions 16x9 \
  --export pptx \
  --output ./exports \
  --open

CLI Options

Option	Short	Description
`--topic`	`-t`	Topic or content to generate
`--file`	`-f`	Path to content file
`--format`		presentation, document, social, webpage
`--slides`	`-n`	Number of slides (1-75)
`--tone`		Tone description
`--audience`		Target audience
`--language`	`-l`	Language code (en, es, pt...)
`--image-model`		AI model (flux-quick, dalle3...)
`--image-style`		Image style description
`--dimensions`	`-d`	Card dimensions
`--export`	`-e`	Export format (pptx, pdf)
`--output`	`-o`	Output directory (default: ./exports)
`--open`		Auto-open exported file after generation
`--quiet`	`-q`	Suppress progress messages
`--timeout`		Generation timeout in seconds (default: 180)
`--help`	`-h`	Show help

Best UX: The `--open` Flag

The --open flag provides seamless workflow — generate and review in one command:

# Create presentation and open immediately
node scripts/gamma-generator.js \
  --file content.md \
  --export pptx \
  --open

# Works on all platforms:
# - Windows: Opens in PowerPoint
# - macOS: Opens in Keynote/PowerPoint
# - Linux: Opens with default application

Two-Step Workflow: Draft → Edit → Generate

For full control over content, use the draft workflow:

Step 1: Generate a Draft Template

# Create editable markdown template (no API call, no credits)
node scripts/gamma-generator.js \
  --topic "AI Ethics for Developers" \
  --slides 10 \
  --tone "thoughtful and practical" \
  --audience "software engineers" \
  --image-style "modern illustrations" \
  --draft \
  --open

This creates a markdown file with:

Slide structure placeholders
Your tone/audience settings preserved
Image style guidance
Opens immediately for editing

Step 2: Edit the Markdown

Fill in your actual content:

## Slide 1: Title
**AI Ethics: A Developer's Responsibility**

Building AI that serves humanity

*Illustration: Developer at computer with ethical symbols*

---

## Slide 2: Why This Matters
**The Code We Write Has Consequences**

- AI systems affect millions of lives
- Bias in training data becomes bias in decisions
- We are the last line of defense

*Illustration: Ripple effect from code to society*

Step 3: Generate from Your Edited File

# Now generate the real presentation
node scripts/gamma-generator.js \
  --file ./exports/ai-ethics-for-developers-draft.md \
  --image-model ideogram \
  --export pptx \
  --open

Why use the draft workflow?

✅ Full control over every slide's content
✅ No wasted credits on iterations
✅ Review structure before committing to generation
✅ Reuse templates for similar presentations

CLI Options Summary

Option	Short	Description
`--topic`	`-t`	Topic or content to generate
`--file`	`-f`	Path to content file
`--format`		presentation, document, social, webpage
`--slides`	`-n`	Number of slides (1-75)
`--tone`		Tone description
`--audience`		Target audience
`--language`	`-l`	Language code (en, es, pt...)
`--image-model`		AI model (flux-quick, dalle3...)
`--image-style`		Image style description
`--dimensions`	`-d`	Card dimensions
`--export`	`-e`	Export format (pptx, pdf)
`--output`	`-o`	Output directory (default: ./exports)
`--open`		Auto-open exported file after generation
`--draft`		Generate markdown template only (no API call)
`--draft-output`		Custom path for draft markdown file
`--quiet`	`-q`	Suppress progress messages
`--timeout`		Generation timeout in seconds (default: 180)
`--help`	`-h`	Show help

Workflow tip: Create a markdown file with your content structure, then generate and review:

# Quick workflow
node scripts/gamma-generator.js -t "My Topic" --draft --open  # Edit the draft
node scripts/gamma-generator.js -f exports/my-topic-draft.md -e pptx --open  # Generate

Run node scripts/gamma-generator.js --help for full documentation.

Related Skills

prompt-engineering - Crafting effective inputText
markdown-mermaid - Diagrams for presentations
image-handling - Working with generated images

Gamma Presentations Skill

Gamma Presentations Skill

When to Use

Prerequisites

Help & User Manual

How to Ask Alex to Create Gamma Content

From a Topic (Alex writes all content)

From Your Own Content File

From Reference Documents

With Specific Options

Command Quick Reference

Available Options

Step-by-Step: Create from a Workspace File

Customization Examples

What Alex Does Behind the Scenes

Pro Tips for Great Results

1. Structure Your Content with Slide Markers

2. Use Illustrations for Professional Decks

3. Match Tone to Audience

4. One Command, Open Result

5. Iterate with Cost-Effective Models

Cost Awareness

Quick Reference

API Base URL

Authentication

Core Endpoints

Generate API Parameters

Required Parameters

Format Options

Text Options

Image Options

Card Dimensions

Export Options

AI Image Models

Cost-Effective (2 credits/image)

Standard (8-15 credits/image)

Premium (20-33 credits/image)

Ultra (30-120 credits/image, Ultra plan only)

Credit Costs

Example Requests

Simple Presentation

Professional Pitch Deck

Document from Notes

Social Media Carousel

Response Handling

POST Response

GET Status Response (Pending)

GET Status Response (Completed)

Polling Pattern

MCP Integration

Setup in Claude

MCP Tools Available

Effective MCP Prompts

Troubleshooting

Insufficient Credits

Authentication Failed

Generation Timeout

Token Limit Exceeded

Best Practices

CLI Script

Quick Examples

CLI Options

Best UX: The --open Flag

Two-Step Workflow: Draft → Edit → Generate

Step 1: Generate a Draft Template

Step 2: Edit the Markdown

Step 3: Generate from Your Edited File

CLI Options Summary

Related Skills

Best UX: The `--open` Flag