runwayml / runway-studio-skills

A skill package that teaches your agent 1 capability — every one documented and browsable below, no GitHub required · by runwayml.

Each skill below is one capability this package teaches your agent. Install the whole package, or open a skill to install just that one.

runway-studio-skills🔥🔥🔥✓ VerifiedFreeQuick setup

Use this skill when the user wants to generate videos, images, or audio using the Runway API. Covers product ad videos, text-to-video, image-to-video, video-to-video, character performance, image generation, sound effects, TTS, speech-to-speech, voice dubbing, and voice isolation.

→

Runway Studio Skills

Generate studio-quality videos, images, and audio with the Runway API at scale. Works with Claude Code, Cursor, Copilot, Codex, Windsurf, and other AI agents.

Install

Agent Skills

npx skills add runwayml/runway-studio-skills

Works with Claude Code, Cursor, Copilot, Codex, Windsurf, and other agents.

Manual

git clone https://github.com/runwayml/runway-studio-skills.git
export RUNWAYML_API_SECRET="your-api-key"

Then ask your agent: "Generate a product ad video from this image"

Setup

Get your API key at dev.runwayml.com
Set RUNWAYML_API_SECRET environment variable (or pass --api-key to any script)
Ensure uv is installed (curl -LsSf https://astral.sh/uv/install.sh | sh) — it ships with Claude Code

Available Scripts

Script	Description
`scripts/generate_video.py`	Text-to-video, image-to-video, video-to-video, character performance
`scripts/generate_image.py`	Text/image to image generation (gen4_image, gen4_image_turbo, gemini_2.5_flash)
`scripts/generate_audio.py`	Sound effects, TTS, speech-to-speech, voice dubbing, voice isolation
`scripts/get_task.py`	Poll, retrieve, or delete tasks
`scripts/runway_helpers.py`	Shared helpers (API client, polling, downloads, error handling)

All scripts use uv run (inline dependencies, no install needed).

Quick Examples

# Generate a product ad video from an image
uv run scripts/generate_video.py \
  --prompt "Smooth camera orbit around the product, soft studio lighting" \
  --image "https://example.com/product.jpg" \
  --model seedance2 --ratio 720:1280 --duration 10

# Generate a video from text
uv run scripts/generate_video.py \
  --prompt "A product bottle rotating slowly on a marble surface" \
  --model seedance2 --ratio 1280:720

# Generate an image
uv run scripts/generate_image.py \
  --prompt "Minimal skincare bottle on white background" \
  --model gen4_image --ratio 1080:1080

# Generate a sound effect
uv run scripts/generate_audio.py sound-effect \
  --prompt "Soft whoosh transition sound" --duration 3

# Text to speech
uv run scripts/generate_audio.py tts \
  --prompt "Welcome to our product showcase" --voice Leslie

# Check task status
uv run scripts/get_task.py --task-id "your-task-uuid" --poll

Documentation

SKILL.md — Full agent skill reference (models, parameters, ratios, error handling)
COOKBOOK.md — End-to-end recipes for product ads and campaigns

API Key

Get your key at dev.runwayml.com. Set as RUNWAYML_API_SECRET environment variable or pass --api-key to any script.

License

MIT

Install the whole package (1 skills):

Copy & paste — that's it

npx skills add https://github.com/runwayml/runway-studio-skills

Or install a single skill:

Copy & paste — that's it

npx skills add https://github.com/runwayml/runway-studio-skills --skill <name>

Pick the skill name from the Skills tab — each entry there installs independently.