Gemini Video Understanding

Gemini Video Understanding Skill

This skill enables comprehensive video analysis using Google's Gemini API, including video summarization, question answering, transcription, timestamp references, and more.

Capabilities

• Video Summarization: Create concise summaries of video content
• Question Answering: Answer specific questions about video content
• Transcription: Transcribe audio with visual descriptions and timestamps
• Timestamp References: Query specific moments in videos (MM:SS format)
• Video Clipping: Process specific segments using start/end offsets
• Multiple Videos: Compare and analyze up to 10 videos (Gemini 2.5+)
• YouTube Support: Analyze YouTube videos directly (preview feature)
• Custom Frame Rate: Adjust FPS sampling for different video types

Supported Formats

• MP4, MPEG, MOV, AVI, FLV, MPG, WebM, WMV, 3GPP

Models Available

Gemini 2.5 Series:

• gemini-2.5-pro - Best quality, 1M context
• gemini-2.5-flash - Balanced quality/speed, 1M context
• gemini-2.5-flash-preview-09-2025 - Preview features, 1M context

Gemini 2.0 Series:

• gemini-2.0-flash - Fast processing
• gemini-2.0-flash-lite - Lightweight option

Context Windows:

• 2M token models: ~2 hours (default) or ~6 hours (low-res)
• 1M token models: ~1 hour (default) or ~3 hours (low-res)

API Key Configuration

The skill supports both Google AI Studio and Vertex AI endpoints.

Option 1: Google AI Studio (Default)

The skill checks for GEMINI_API_KEY in this order: 1. Process environment: process.env.GEMINI_API_KEY or $GEMINI_API_KEY 2. Project root: .env 3. .claude directory: .claude/.env 4. .claude/skills directory: .claude/skills/.env 5. Skill directory: .claude/skills/gemini-video-understanding/.env

Get your API key: https://aistudio.google.com/apikey

To set up:

# Environment variable (recommended)
export GEMINI_API_KEY="your-api-key-here"

# Or in .env file
echo "GEMINI_API_KEY=your-api-key-here" > .env

Option 2: Vertex AI

To use Vertex AI instead:

# Enable Vertex AI
export GEMINI_USE_VERTEX=true
export VERTEX_PROJECT_ID=your-gcp-project-id
export VERTEX_LOCATION=us-central1  # Optional, defaults to us-central1

Or in .env file:

GEMINI_USE_VERTEX=true
VERTEX_PROJECT_ID=your-gcp-project-id
VERTEX_LOCATION=us-central1

Usage Instructions

When to Use This Skill

Use this skill when the user asks to:

• Analyze, summarize, or describe video content
• Answer questions about videos
• Transcribe video audio with visual context
• Extract information from specific timestamps
• Compare multiple videos
• Process YouTube video content
• Create quizzes or educational content from videos

Basic Video Analysis

For video files:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/video.mp4" \
  --prompt "Summarize this video in 3 key points"

For YouTube URLs:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --youtube-url "https://www.youtube.com/watch?v=VIDEO_ID" \
  --prompt "What are the main topics discussed?"

Advanced Features

Video Clipping (specific time range):

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/video.mp4" \
  --prompt "Summarize this segment" \
  --start-offset "40s" \
  --end-offset "80s"

Custom Frame Rate:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/video.mp4" \
  --prompt "Analyze the rapid movements" \
  --fps 5

Transcription with Timestamps:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/video.mp4" \
  --prompt "Transcribe the audio with timestamps and visual descriptions"

Multiple Videos (Gemini 2.5+ only):

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-paths "/path/video1.mp4" "/path/video2.mp4" \
  --prompt "Compare these two videos and highlight the differences"

Model Selection:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/video.mp4" \
  --prompt "Detailed analysis" \
  --model "gemini-2.5-pro"

Script Parameters

Required (one of):
  --video-path PATH           Path to local video file
  --youtube-url URL           YouTube video URL
  --video-paths PATH [PATH..] Multiple video paths (Gemini 2.5+)

Required:
  --prompt TEXT              Analysis prompt/question

Optional:
  --model NAME               Model to use (default: gemini-2.5-flash)
  --start-offset TIME        Video clip start (e.g., "40s", "1m30s")
  --end-offset TIME          Video clip end (e.g., "80s", "2m")
  --fps NUMBER               Frame sampling rate (default: 1)
  --output-file PATH         Save response to file
  --verbose                  Show detailed processing info

Common Use Cases

1. Video Summarization

Prompt: "Summarize this video in 3 key points with timestamps"

2. Educational Content

Prompt: "Create a quiz with 5 questions and answer key based on this video"

3. Timestamp-Specific Questions

Prompt: "What happens at 01:15 and how does it relate to the topic at 02:30?"

4. Transcription

Prompt: "Transcribe the audio from this video with timestamps for salient events and visual descriptions"

5. Content Comparison

Prompt: "Compare these two product demo videos. Which one explains the features more clearly?"

6. Action Detection

Prompt: "List all the actions performed in this tutorial video with timestamps"

Rate Limits & Quotas

Free Tier (per model):

• 10-15 RPM (requests per minute)
• 1M-4M TPM (tokens per minute)
• 1,500 RPD (requests per day)

YouTube Limitations:

• Free tier: 8 hours of YouTube video per day
• Paid tier: No length-based limits
• Public videos only (no private/unlisted)

Storage (Files API):

• 20GB per project
• 2GB per file
• 48-hour retention period

Token Calculation

Video tokens depend on resolution:

• Default resolution: ~300 tokens per second of video
• Low resolution: ~100 tokens per second of video

Example: A 10-minute video = 600 seconds × 300 tokens = ~180,000 tokens

Error Handling

Common errors and solutions:

Best Practices

1. Use Files API for videos >20MB - More reliable than inline data 2. Wait for file processing - Poll until state is ACTIVE before analysis 3. Optimize FPS - Use lower FPS for static content to save tokens 4. Clip long videos - Process specific segments instead of entire video 5. Cache context - Reuse uploaded files for multiple queries 6. Batch processing - Process multiple short videos in one request (2.5+) 7. Specific prompts - Be precise about what you want to extract

Implementation Notes

For Claude Code:

When a user requests video analysis:

1. Check API key availability first using the helper script 2. Determine video source: local file, YouTube URL, or multiple videos 3. Select appropriate model based on requirements (default: gemini-2.5-flash) 4. Run the analysis script with proper parameters 5. Parse and present results to the user clearly 6. Handle errors gracefully with helpful suggestions

Files API Workflow:

For videos >20MB or reusable content: 1. Upload video using Files API (script handles this automatically) 2. Wait for ACTIVE state (polling included in script) 3. Use file URI for analysis 4. Files auto-delete after 48 hours

Inline Data Workflow:

For videos <20MB: 1. Read video file as bytes 2. Base64 encode for API 3. Send in generateContent request 4. Single-use, no upload needed

Example Workflows

Workflow 1: YouTube Video Summary

# User: "Analyze this YouTube tutorial video"
python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --youtube-url "https://www.youtube.com/watch?v=abc123" \
  --prompt "Create a structured summary with: 1) Main topics, 2) Key takeaways, 3) Recommended audience"

Workflow 2: Interview Transcription

# User: "Transcribe this interview with timestamps"
python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "interview.mp4" \
  --prompt "Transcribe this interview with speaker labels, timestamps, and visual descriptions of gestures or slides shown"

Workflow 3: Product Comparison

# User: "Compare these two product demo videos"
python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-paths "demo1.mp4" "demo2.mp4" \
  --model "gemini-2.5-pro" \
  --prompt "Compare these product demos on: features shown, presentation quality, clarity of explanation, and overall effectiveness"

Troubleshooting

API Key Not Found:

# Check API key detection
python .claude/skills/gemini-video-understanding/scripts/check_api_key.py

Video Too Large:

Error: Request size exceeds 20MB
Solution: Script automatically uses Files API for large videos

Processing Timeout:

Error: File not reaching ACTIVE state
Solution: Check video integrity, try smaller file, or different format

Rate Limit Errors:

Error: 429 Too Many Requests
Solution: Wait before retry, or upgrade to paid tier

Additional Resources

• API Documentation: https://ai.google.dev/gemini-api/docs/video-understanding
• Files API Guide: https://ai.google.dev/gemini-api/docs/vision#uploading-files
• Rate Limits: https://ai.google.dev/gemini-api/docs/rate-limits
• Pricing: https://ai.google.dev/pricing
• Get API Key: https://aistudio.google.com/apikey

Version History

• 1.0.0 (2025-10-26): Initial release with full video understanding capabilities

Gemini Video Understanding - Examples

Setup

1. Install dependencies:

pip install -r .claude/skills/gemini-video-understanding/requirements.txt

2. Configure API key:

# Option 1: Environment variable (recommended)
export GEMINI_API_KEY="your-api-key-here"

# Option 2: Copy and edit .env file
cp .claude/skills/gemini-video-understanding/.env.example \
   .claude/skills/gemini-video-understanding/.env
# Then edit the .env file with your API key

3. Verify configuration:

python .claude/skills/gemini-video-understanding/scripts/check_api_key.py

Basic Examples

Example 1: Simple Video Summary

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/video.mp4" \
  --prompt "Summarize this video in 3 key points"

Output:

1. The video demonstrates how to use the Gemini API for video analysis
2. It covers three main input methods: Files API, inline data, and YouTube URLs
3. Examples show various use cases including transcription and timestamp queries

Example 2: YouTube Video Analysis

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --youtube-url "https://www.youtube.com/watch?v=dQw4w9WgXcQ" \
  --prompt "What is this video about? List the main topics discussed."

Example 3: Detailed Transcription

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/interview.mp4" \
  --prompt "Transcribe this interview with timestamps and speaker labels" \
  --verbose

Advanced Examples

Example 4: Video Clipping

Analyze only a specific portion of a video:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/long-video.mp4" \
  --prompt "Summarize this segment" \
  --start-offset "2m30s" \
  --end-offset "5m15s"

Example 5: High Frame Rate Analysis

For videos with fast action:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/sports.mp4" \
  --prompt "Describe the key movements and techniques shown" \
  --fps 5

Example 6: Comparing Multiple Videos

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-paths "/path/video1.mp4" "/path/video2.mp4" \
  --prompt "Compare these two product demos. Which one is more effective and why?" \
  --model "gemini-2.5-pro"

Example 7: Educational Content Generation

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --youtube-url "https://www.youtube.com/watch?v=VIDEO_ID" \
  --prompt "Create a quiz with 5 multiple-choice questions based on this video. Include an answer key with explanations."

Example 8: Timestamp-Specific Questions

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/tutorial.mp4" \
  --prompt "What happens at 01:15 and 02:30? How are these two moments related?"

Prompting Best Practices

For Summaries

"Provide a structured summary with:
1. Main topic or theme
2. Key points discussed (3-5 bullet points)
3. Notable quotes or moments with timestamps
4. Target audience or intended purpose"

For Transcriptions

"Transcribe this video including:
- Speaker labels (Speaker 1, Speaker 2, etc.)
- Timestamps for each segment (MM:SS format)
- Visual descriptions in [brackets] when relevant
- Key gestures or on-screen text mentioned"

For Educational Content

"Analyze this educational video and create:
1. Learning objectives (3-5 points)
2. Key concepts explained with timestamps
3. 5 quiz questions with multiple choice answers
4. Answer key with brief explanations"

For Comparisons

"Compare these videos on the following criteria:
1. Content quality and accuracy
2. Presentation style and clarity
3. Production quality (audio, video, editing)
4. Target audience suitability
5. Overall effectiveness
Provide ratings (1-5) for each criterion with justification."

Use Case Examples

Use Case 1: Meeting Notes

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "team-meeting.mp4" \
  --prompt "Create meeting notes with: 1) Attendees (if visible), 2) Topics discussed with timestamps, 3) Action items identified, 4) Decisions made" \
  --output-file "meeting-notes.txt"

Use Case 2: Content Moderation

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "user-submitted.mp4" \
  --prompt "Analyze this video for: 1) Overall content type, 2) Any inappropriate content, 3) Compliance with community guidelines, 4) Recommended action (approve/review/reject)"

Use Case 3: Accessibility - Video Descriptions

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "promotional-video.mp4" \
  --prompt "Create an audio description track for visually impaired viewers. Include descriptions of: visual elements, on-screen text, scene changes, and important actions at appropriate timestamps."

Use Case 4: Sports Analysis

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "game-footage.mp4" \
  --prompt "Analyze this game footage: 1) Key plays with timestamps, 2) Player performance highlights, 3) Strategic decisions, 4) Turning points in the game" \
  --fps 2 \
  --verbose

Use Case 5: Tutorial Enhancement

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --youtube-url "https://www.youtube.com/watch?v=TUTORIAL_ID" \
  --prompt "Create an enhanced tutorial outline with: 1) Chapter markers with timestamps, 2) Prerequisites mentioned, 3) Tools/resources needed, 4) Common mistakes to avoid, 5) Practice exercises suggested"

Output Options

Save to File

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "video.mp4" \
  --prompt "Summarize this video" \
  --output-file "summary.txt"

JSON Output

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "video.mp4" \
  --prompt "Summarize this video" \
  --json \
  --output-file "summary.json"

Verbose Mode (with token usage)

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "video.mp4" \
  --prompt "Analyze this video" \
  --verbose

Error Handling

Check API Key

python .claude/skills/gemini-video-understanding/scripts/check_api_key.py

Common Errors

File Not Found:

# Error: Video file not found: /path/to/video.mp4
# Solution: Check file path and ensure file exists
ls -lh /path/to/video.mp4

API Key Invalid:

# Error: 403 Forbidden
# Solution: Verify API key is correct
python .claude/skills/gemini-video-understanding/scripts/check_api_key.py

Rate Limit:

# Error: 429 Too Many Requests
# Solution: Wait before retrying, or upgrade to paid tier

Model Selection

Fast Processing

--model "gemini-2.0-flash-lite"  # Fastest, good for simple tasks

Balanced (Default)

--model "gemini-2.5-flash"  # Best balance of speed and quality

Highest Quality

--model "gemini-2.5-pro"  # Most accurate, slower, higher token usage

Performance Tips

1. Use video clipping to analyze only relevant portions 2. Adjust FPS based on content type (lower for static, higher for action) 3. Use Files API for videos >20MB (handled automatically) 4. Batch process multiple short videos in one request (Gemini 2.5+) 5. Cache uploaded files - reuse file URIs for multiple queries

Resources

• API Documentation: https://ai.google.dev/gemini-api/docs/video-understanding
• Get API Key: https://aistudio.google.com/apikey
• Pricing: https://ai.google.dev/pricing
• Rate Limits: https://ai.google.dev/gemini-api/docs/rate-limits

Quick Start Guide - Gemini Video Understanding

1. Install Dependencies

pip install google-genai

Or use the requirements file:

pip install -r .claude/skills/gemini-video-understanding/requirements.txt

2. Configure API Key

Choose one option:

Option A: Environment Variable (Recommended)

export GEMINI_API_KEY="your-api-key-here"

Option B: Skill Directory .env

cp .claude/skills/gemini-video-understanding/.env.example \
   .claude/skills/gemini-video-understanding/.env
# Edit the .env file with your API key

Option C: Project Root .env

echo "GEMINI_API_KEY=your-api-key-here" > .env

Get your API key at: https://aistudio.google.com/apikey

3. Verify Setup

python .claude/skills/gemini-video-understanding/scripts/check_api_key.py

4. Analyze Your First Video

Local video file:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "/path/to/video.mp4" \
  --prompt "Summarize this video in 3 sentences"

YouTube video:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --youtube-url "https://www.youtube.com/watch?v=VIDEO_ID" \
  --prompt "What are the main topics discussed?"

5. Common Commands

Transcribe with timestamps:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "video.mp4" \
  --prompt "Transcribe this video with timestamps and visual descriptions"

Analyze specific segment:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "video.mp4" \
  --prompt "Summarize this part" \
  --start-offset "1m30s" \
  --end-offset "3m"

Compare videos:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-paths "video1.mp4" "video2.mp4" \
  --prompt "Compare these videos"

Save to file:

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py \
  --video-path "video.mp4" \
  --prompt "Analyze this video" \
  --output-file "analysis.txt"

Help

python .claude/skills/gemini-video-understanding/scripts/analyze_video.py --help

See EXAMPLES.md for more detailed examples and use cases.

Gemini Video Understanding Skill

A comprehensive Claude Code skill for analyzing videos using Google's Gemini API.

Overview

This skill enables AI-powered video analysis with capabilities including:

• Video summarization and content description
• Question answering about video content
• Audio transcription with visual descriptions
• Timestamp-based queries (MM:SS format)
• Video clipping (specific time ranges)
• Multiple video comparison (Gemini 2.5+)
• YouTube video processing
• Custom frame rate sampling

Features

✅ Three Video Input Methods:

• Local files (Files API for >20MB, inline for <20MB)
• YouTube URLs (preview feature)
• Multiple videos (up to 10 with Gemini 2.5+)

✅ Advanced Processing:

• Video clipping with start/end offsets
• Custom FPS sampling (default: 1 FPS)
• Context windows up to 2M tokens (~6 hours of video)

✅ 9 Supported Video Formats: MP4, MPEG, MOV, AVI, FLV, MPG, WebM, WMV, 3GPP

✅ Multiple Models:

• Gemini 2.5 Series (Pro, Flash, Preview)
• Gemini 2.0 Series (Flash, Flash-Lite)

✅ Flexible API Key Management: Checks in order: Process env → Skill directory → Project root

Quick Start

1. Install: pip install google-genai 2. Configure: export GEMINI_API_KEY="your-key" 3. Verify: python scripts/check_api_key.py 4. Analyze: python scripts/analyze_video.py --video-path video.mp4 --prompt "Summarize"

See QUICKSTART.md for detailed setup instructions.

Documentation

• [SKILL.md](SKILL.md) - Complete skill documentation
• [QUICKSTART.md](QUICKSTART.md) - Quick setup guide
• [EXAMPLES.md](EXAMPLES.md) - Comprehensive examples and use cases

Directory Structure

gemini-video-understanding/
├── SKILL.md              # Main skill documentation
├── README.md             # This file
├── QUICKSTART.md         # Quick start guide
├── EXAMPLES.md           # Detailed examples
├── requirements.txt      # Python dependencies
├── .env.example          # API key template
└── scripts/
    ├── analyze_video.py  # Main video analysis script
    └── check_api_key.py  # API key verification tool

Common Use Cases

API Key Configuration

The skill automatically checks for GEMINI_API_KEY in this order:

1. Process Environment - $GEMINI_API_KEY 2. Skill Directory - .claude/skills/gemini-video-understanding/.env 3. Project Root - .env file

Get your API key at: https://aistudio.google.com/apikey

Rate Limits & Pricing

Free Tier:

• 10-15 requests per minute
• 1-4M tokens per minute
• 1,500 requests per day
• YouTube: 8 hours per day

Token Usage:

• Default: ~300 tokens/second of video
• Low-res: ~100 tokens/second of video

See full pricing: https://ai.google.dev/pricing

Support

• API Documentation: https://ai.google.dev/gemini-api/docs/video-understanding
• Files API: https://ai.google.dev/gemini-api/docs/vision#uploading-files
• Rate Limits: https://ai.google.dev/gemini-api/docs/rate-limits
• Issue Tracker: [Report issues on GitHub]

License

MIT License - See skill metadata for details.

Version

1.0.0 (2025-10-26)