Sora Video Generation

Generate videos from text prompts or reference images using OpenAI Sora. ✅ USE WHEN: - Need AI-generated video from text description - Want image-to-video (animate a still image) - Creating cinematic/artistic video content - Need motion/animation without lip-sync ❌ DON'T USE WHEN: - Need lip-sync (person speaking) → use veed-ugc or ugc-manual - Just need image generation → use nano-banana-pro or morpheus - Editing existing videos → use Remotion - Need UGC-style talking head → use veed-ugc INPUT: Text prompt + optional reference image OUTPUT: MP4 video (various resolutions/durations)

Security Scan

The skill appears to do what it claims (call OpenAI Sora to generate videos), but there are a few inconsistencies you should understand before installing — most importantly the skill uses an OPENAI_API_KEY but the metadata does not declare required credentials and there is no install spec for declared Python dependencies.

Purpose & Capability

Name, description, and runtime behavior align: the script calls OpenAI's videos endpoints to create and download Sora-generated videos. However the registry metadata lists no required environment variables even though the SKILL.md and script clearly require an OpenAI API key (OPENAI_API_KEY or --api-key).

Instruction Scope

SKILL.md instructions and the script stay on-topic: they take a prompt and optional image, resize the image, call the OpenAI Videos API, poll for completion, and download the MP4. The instructions do not request unrelated files, system credentials, or unexpected external endpoints.

Install Mechanism

This is instruction-only / no install spec, so nothing is auto-downloaded by an installer. The included Python script however lists dependencies (openai, httpx, pillow) but the skill does not provide an installation step. The user will need to ensure those packages and Python >=3.10 are installed before running.

Credentials

The script requires an OpenAI API key (OPENAI_API_KEY or --api-key) but the skill metadata declares no required env vars or primary credential. That omission is a mismatch and could lead to confusion; otherwise the script does not request unrelated credentials or broad system secrets.

Persistence & Privilege

The skill does not request permanent/autonomous privileges (always:false) and does not modify other skills or system-wide agent configuration. It writes output files and temporary image files only as needed for its function.

What to consider before installing

This skill is functionally coherent with its description (it calls OpenAI Sora to generate videos), but take these precautions before installing or running it: - Provide only a scoped OpenAI API key and be aware this key will be used to create video jobs and download content; verify billing and key permissions. The skill uses OPENAI_API_KEY (or --api-key) but the registry metadata omitted this — expect to supply it yourself. - Install the required Python environment and libraries (Python >=3.10, openai, httpx, pillow) or run in a controlled environment (virtualenv/venv) to avoid affecting system packages. - The code saves temp image files and the final MP4 to disk; ensure you run it in a directory where writing files is acceptable. - Videos may expire on the provider side (~1 hour), so they are downloaded immediately by the script; be mindful of any sensitive content sent to the API. - The skill source and homepage are unknown — if you need higher assurance, ask the publisher for provenance (where the package came from, signed releases, or an official repo) before trusting it with an API key. If you want to proceed, consider creating a dedicated OpenAI API key with limited scope or billing limits to reduce risk.

Generate videos using OpenAI's Sora API.

API Reference

Endpoint: POST https://api.openai.com/v1/videos

Parameters

Parameter	Values	Description
prompt	string	Text description of the video (required)
input_reference	file	Optional image that guides generation
model	sora-2, sora-2-pro	Model to use (default: sora-2)
seconds	4, 8, 12	Video duration (default: 4)
size	720x1280, 1280x720, 1024x1792, 1792x1024	Output resolution

Important Notes

• Image dimensions must match video size exactly - the script auto-resizes
• Video generation takes 1-3 minutes typically
• Videos expire after ~1 hour - download immediately

Usage

# Basic text-to-video
uv run ~/.clawdbot/skills/sora/scripts/generate_video.py \
  --prompt "A cat playing piano" \
  --filename "output.mp4"

# Image-to-video (auto-resizes image)
uv run ~/.clawdbot/skills/sora/scripts/generate_video.py \
  --prompt "Slow dolly shot, steam rising, warm lighting" \
  --filename "output.mp4" \
  --input-image "reference.png" \
  --seconds 8 \
  --size 720x1280

# With specific model
uv run ~/.clawdbot/skills/sora/scripts/generate_video.py \
  --prompt "Cinematic scene" \
  --filename "output.mp4" \
  --model sora-2-pro \
  --seconds 12

Script Parameters

Flag	Description	Default
--prompt, -p	Video description (required)	-
--filename, -f	Output file path (required)	-
--input-image, -i	Reference image path	None
--seconds, -s	Duration: 4, 8, or 12	8
--size, -sz	Resolution	720x1280
--model, -m	sora-2 or sora-2-pro	sora-2
--api-key, -k	OpenAI API key	env var
--poll-interval	Check status every N seconds	10

API Key

Set OPENAI_API_KEY environment variable or pass --api-key.

Prompt Engineering for Video

Good prompts include:

1. Camera movement: dolly, pan, zoom, tracking shot
2. Motion description: swirling, rising, falling, shifting
3. Lighting: golden hour, candlelight, dramatic rim lighting
4. Atmosphere: steam, particles, bokeh, haze
5. Mood/style: cinematic, commercial, lifestyle, editorial

Example prompts:

Food commercial:

Slow dolly shot of gourmet dish, soft morning sunlight streaming through window, 
subtle steam rising, warm cozy atmosphere, premium food commercial aesthetic

Lifestyle:

Golden hour light slowly shifting across mountains, gentle breeze rustling leaves, 
serene morning atmosphere, premium lifestyle commercial

Product shot:

Cinematic close-up, dramatic lighting with warm highlights, 
slow reveal, luxury commercial style

Workflow: Image → Video

1. Generate image with Nano Banana Pro (or use existing)
2. Pass image as --input-image to Sora
3. Write prompt describing desired motion/atmosphere
4. Script auto-resizes image to match video dimensions

Output

• Videos saved as MP4
• Typical file size: 1.5-3MB for 8 seconds
• Resolution matches --size parameter