Skip to main content
skill CX Codex

scenario-luma-video — Claude Code skill: what it does, how to install

Generates and edits video using Luma Ray models on Scenario via MCP. It routes prompts to the right model for text-to-video, restyling, or outpainting new aspect ratios.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add scenario-labs/skills --skill scenario-luma-video -a claude-code
Open on GitHub

What scenario-luma-video does

This skill routes video tasks across four distinct Luma models on Scenario via MCP. Ray 3.2 generates clips from text prompts or anchor frames. Ray 3.2 Edit restyles footage while preserving movement using pose, depth, and trajectory controls. Ray 3.2 Reframe outpaces letterboxing by generating new borders rather than cropping the subject. Modify Video handles lighter, budget-friendly prompt edits on clips up to 30 seconds and 100 MB.

Working with this skill requires Scenario MCP tools and the companion scenario skill for connection and asset uploads. Model parameters strictly enforce specific restrictions. For instance, duration must be passed as "5s" or "10s", and "10s" completely rejects looping, HDR, and frame anchors. Both editors share a nine-step transformation ladder from adhere_1 to reimagine_3.

Ray 3.2 Edit ran nearly two orders of magnitude more expensive than Modify Video at authoring time. Because of this massive cost gap, testing parameters with dry_run=true on model_run is essential before submitting batches. Prompts should focus on physical action and camera movement rather than decorative adjectives.

What it helps with

  • generate videos from natural language prompts or anchor frames in Ray 3.2
  • restyle existing footage while tracking pose, depth, and motion trajectory
  • expand video canvases to new aspect ratios using outpainting without cropping
  • apply budget-friendly prompt adjustments to clips under 30 seconds

Example prompts for the agent

  • Generate a 5s 16:9 clip with Ray 3.2 using asset_1 as startFrame and a slow push-in camera move
  • Reframe this video to 9:16 using Ray 3.2 Reframe, filling the expanded background with soft lighting
  • Restyle footage via Ray 3.2 Edit with adhere_2 strength and trajectory tracking enabled

How to install scenario-luma-video

1Skills CLI (Vercel)

npx skills add scenario-labs/skills --skill scenario-luma-video -a claude-code

2Manually: put the files in place

~/.claude/skills/scenario-luma-video/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/scenario-luma-video/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about scenario-luma-video

How to install scenario-luma-video in Claude Code?

Run npx skills add scenario-labs/skills --skill scenario-luma-video -a claude-code.

What does the scenario-luma-video skill do?

It interfaces with Luma Ray models on Scenario via MCP, handling video generation, motion-preserving restyling, outpainting reframes, and budget clip editing.

What are the restrictions on 10s clips in Ray 3.2?

Ray 3.2 rejects loop, hdr, startFrame, and endFrame whenever duration is set to "10s".

Source text

SKILL.md · MIT
---
name: scenario-luma-video
description: "Use when generating or editing video with Luma Ray models on Scenario via MCP: cinematic text-to-video, image-to-video from a start frame, start plus end frame anchors, seamless looping clips, HDR output, restyling footage while preserving motion with edit strengths and face, pose, depth, or trajectory controls, reframing to another aspect ratio by outpainting, or budget prompt edits on real footage. Keywords: Luma, Dream Machine, Ray 3.2, Ray 3, Modify Video, Reframe, T2V, I2V, V2V, loop, HDR."
license: MIT
---

# Scenario Luma Video

## Overview

Luma's video line on Scenario is four sibling models, one per mode: Ray 3.2 generates from text or
frame anchors, Ray 3.2 Edit restyles footage under structural controls, Ray 3.2 Reframe outpaints to
a new aspect ratio, and Modify Video makes budget prompt edits. Route by member before tuning
anything; Luma's image models belong to the `scenario-luma-image` skill. Discover the live set with
`search` and treat `model_schema_get` as the contract.

Connection and the core loop: see the `scenario` skill in this repo; model-agnostic video work: the
`scenario-video` skill. If a sibling skill named here is missing from your available skills, ask the
user to install it (`npx skills add scenario-labs/skills --skill <name>`); unattended, proceed from
tool schemas and flag the gap.

## Quick reference

Each mode is its own model (names from the live schemas, caps as of authoring time):

| Member          | Inputs                                | Behavior                                                                 |
| --------------- | ------------------------------------- | ------------------------------------------------------------------------ |
| Ray 3.2         | `prompt` (+ `startFrame`, `endFrame`) | text or image to video; `endFrame` is only valid alongside `startFrame`  |
| Ray 3.2 Edit    | `video` + `prompt`                    | restyles while keeping motion and timing; `editStrength` plus controls   |
| Ray 3.2 Reframe | `video` + `prompt` + `aspectRatio`    | fits a new ratio by outpainting, never cropping; prompt fills the canvas |
| Modify Video    | `video` (+ `prompt`, `firstFrame`)    | budget prompt edits via a `mode` ladder; source up to 30 seconds, 100 MB |

On Ray 3.2 the options veto one another. At authoring time `duration` was a unit string, "5s" or
"10s", never a number; "10s" refused `loop`, `hdr`, `startFrame`, and `endFrame`. `hdr` required
720p or 1080p from the 540p, 720p, 1080p `resolution` list; six `aspectRatio` values ran 9:16
through 21:9. Edit and Reframe share that resolution list; Modify Video has no resolution parameter.
`duration`, `resolution`, and `hdr` each move the price (1080p ran several times 720p's cost and
wait), so `dry_run` the option set before a batch.

## Two editors, one strength ladder

Ray 3.2 Edit and Modify Video restyle through the same nine-value ladder, `adhere_1` through
`reimagine_3` (adhere stays close, flex restyles but keeps recognizable elements, reimagine
transforms), and share little else. Edit names the ladder `editStrength`; Modify names it `mode`.
Edit takes a guide image as `startFrame`; Modify's `firstFrame` is an edited copy of the source's
own first frame, not an arbitrary style image. Edit alone offers `keyframes` (up to 64
image-and-index anchors, mutually exclusive with the single `startFrame` guide), face, pose, depth,
normals, and trajectory conditioning toggles, resolution choice, and `hdr`; or pass
`autoControls: true` instead of a manual `editStrength`. The price gap ran near two orders of
magnitude at authoring time, so `dry_run` the same clip on both and pay for Edit's controls only
when the edit needs them.

## Prompt motion, not adjectives

Ray rewards natural-language prompts of roughly 50 to 300 words built around motion: a subject
mid-action, one named camera move (slow push-in, handheld follow), and one physical consequence of
the action (droplets scattering, dust rising). Concrete lighting language lands directly. With
`loop`, hold motion intensity constant (steady rain, drifting steam) so the cycle closes cleanly. On
Reframe, prompt only the edges: describe what the new canvas should contain and leave the original
subject alone.

## Worked example: a 16:9 hero clip recut for 9:16

1. `search` with `target="models"`, `query="luma"`, `public=true`. Prefer the newest non-deprecated
   generation, e.g. `model_luma-ray-3-2` and its Edit and Reframe siblings (live hits at authoring
   time: re-discover each session).
2. `model_schema_get` with the generator id: options and their vetoes before anything else.
3. `upload_asset` the product still (see the `scenario` skill) to get an asset id.
4. `model_run` with that `model_id`, `dry_run=true`, and the exact
   `parameters={"prompt": "A crystal perfume bottle catches soft window light as a single drop arcs off the stopper, slow push-in, product commercial style.", "startFrame": "asset_a", "duration": "5s", "resolution": "1080p", "aspectRatio": "16:9"}`;
   re-estimate after any option change.
5. Repeat `model_run` with `wait=false`, then `jobs_wait` with the returned job id, re-called with
   `pending_job_ids` on timeout, never a second `model_run`.
6. `model_schema_get` the Reframe sibling, then `model_run` with
   `parameters={"video": "<generated asset id>", "aspectRatio": "9:16", "prompt": "continue the marble counter downward and the softly lit wall upward", "resolution": "1080p"}`,
   same wait discipline.
7. `asset_display` both and inspect the outpainted edges before delivery.

## Common mistakes

- `duration: 5` or `"5"`: Ray 3.2 takes the unit string, `"5s"` or `"10s"`.
- A 10 second loop, HDR grade, or frame anchor: `"10s"` excludes all three; drop to `"5s"`.
- Ray parameter names on Modify Video: it takes `firstFrame` and `mode`, not `startFrame` and
  `editStrength`.
- `editStrength` with `autoControls`, or `startFrame` with `keyframes`: each pair is mutually
  exclusive on Edit.
- Prompting the whole scene on Reframe: the prompt describes only the added canvas; the source
  subject stays.
- Vibrant, whimsical, or hyper-realistic in a Ray prompt: they degrade quality in this family.

© scenario-labs · MIT · scenario-labs/skills

Similar skills

Everything for AI and ML →

From the same set68

scenario-minimax-video A skill for generating and editing video with MiniMax Hailuo models on Scenario: text-to-video, frame anchors, clip extension, recasting people. CX Codex +1 375installs scenario-luma-image A skill for generating and editing images with Luma Uni-1 models on Scenario via MCP. Covers posters with exact text, styled scenes, and prompt-based edits. CX Codex +1 375installs scenario-reve Controls Scenario's Reve v2.1 and Reve Remix image models over MCP for prompt-only generation, instruction edits, and multi-image blending. CX Codex +1 367installs scenario-mai-image A skill for driving Microsoft MAI Image models on Scenario over MCP: text-to-image and instruction-based image edits. CX Codex +1 375installs scenario-grok-imagine-video A skill for generating and editing video with Grok Imagine models on Scenario over MCP. Covers text-to-video, first-frame animation, reference-based consistency, prompt editing, and clip extension. CX Codex +1 372installs scenario-orbit-views Builds consistent multi-angle views and 360 turnarounds of a subject from a single image using Blender and Scenario MCP. CX Codex +1 365installs scenario-wan A skill for generating and editing video with Wan models through Scenario MCP: text-to-video, animating a still, instruction-based edits, reframing. CX Codex +1 370installs scenario-blender-hard-surface A Blender hard-surface modeling skill: boolean cuts, bevels, weighted normals and subdivision. Handles broken shading and bevels that vanish after a boolean. CX Codex +1 371installs

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor