neodomain-ai

Generate images and videos via Neodomain AI API. Supports text-to-image, image-to-video, text-to-video, and motion control video generation. Use when user wa...

Security Scan

The skill appears to do what it claims (wrap Neodomain API calls) and requests only a service token and python3, but there are a few operational gaps you should be aware of before installing.

Purpose & Capability

Name/description align with the code and SKILL.md: scripts call Neodomain endpoints (story.neodomain.cn) to generate images/videos and perform authentication. Requiring python3 and NEODOMAIN_ACCESS_TOKEN is appropriate for this purpose.

Instruction Scope

Runtime instructions and scripts stay within the generation workflow, but several scripts (e.g., batch_video.py) upload user files to the provider's OSS (wlpaas.oss-cn-shanghai.aliyuncs.com) via temporary STS credentials. The SKILL.md mentions authentication but does not explicitly warn that user images/storyboards will be uploaded to the provider's OSS; the login flow requires the user to provide a phone/email and verification code (expected for token issuance).

Install Mechanism

This is instruction-only (no platform install spec), which reduces install risk. However, some scripts import third-party Python packages (notably oss2 in batch_video.py) but the SKILL.md / INSTALL.md do not declare or automate these Python dependency installs—users will need to pip-install dependencies manually. There are no external archive downloads or short/unknown URLs used for installing code.

Credentials

Only NEODOMAIN_ACCESS_TOKEN is required and is the primary credential; that is proportional to the stated functionality. The login script collects a contact (phone/email) and verification code from the user to obtain the token, which is consistent with the service's auth flow.

Persistence & Privilege

The skill is not forced-always or privileged; it doesn't request persistent platform privileges or modify other skills. Autonomous invocation is allowed by default (normal for skills) but not excessive here.

Assessment

This package appears to be a straightforward client for the Neodomain service, but check the following before installing or running it: - Understand that images/videos and storyboard files you provide may be uploaded to the provider's OSS (wlpaas.oss-cn-shanghai.aliyuncs.com) via temporary STS credentials—do not send sensitive/private images if you don't trust the service. - You will need to provide NEODOMAIN_ACCESS_TOKEN (or run the login flow which asks for your phone/email and verification code). Keep that token secret and avoid committing it to files or shared shells. - The scripts import a third-party Python library (oss2) but the README does not automatically install Python deps—install required packages (e.g., pip install oss2) in a controlled virtualenv before running. - Review network endpoints (story.neodomain.cn and wlpaas.oss-cn-shanghai.aliyuncs.com) to ensure they are acceptable for your data/privacy policies. If you want higher assurance, ask the maintainer for a full dependency list (requirements.txt) and an explicit statement that user data is only uploaded to the Neodomain service for generation purposes.

Neodomain AI Content Generator

Generate images and videos using the Neodomain AI platform API.

Setup

Set your access token as an environment variable:

export NEODOMAIN_ACCESS_TOKEN="your_access_token_here"

Or pass it directly to scripts via --token flag.

Image Generation

Get Available Models

python3 {baseDir}/scripts/image_models.py --token $NEODOMAIN_ACCESS_TOKEN

Generate Images

# Basic text-to-image
python3 {baseDir}/scripts/generate_image.py --prompt "A futuristic city at sunset" --token $NEODOMAIN_ACCESS_TOKEN

# With options
python3 {baseDir}/scripts/generate_image.py \
  --prompt "A beautiful mountain landscape" \
  --negative-prompt "blurry, low quality" \
  --model "doubao-seedream-4-0" \
  --aspect-ratio "16:9" \
  --num-images 4 \
  --size "2K" \
  --output-dir ./output/images \
  --token $NEODOMAIN_ACCESS_TOKEN

Parameters

Parameter	Description	Default
--prompt	Text description for image generation	Required
--negative-prompt	Things to exclude from image	Empty
--model	Model name (see models list)	gemini-3.1-flash-image-preview
--aspect-ratio	Image aspect ratio: 1:1, 16:9, 9:16, 4:3, 3:4	1:1
--num-images	Number of images to generate: 1 or 4	1
--size	Image size: 1K, 2K, 4K	2K
--guidance-scale	Prompt adherence (1.0-20.0)	7.5
--seed	Random seed for reproducibility	Random
--output-format	Output format: jpeg, png, webp	jpeg
--output-dir	Where to save images	./output

Image Generation with Reference Images

Use generate_image_ref.py to generate images with character reference(s). Supports up to 10 reference images per generation:

# Single reference image
python3 {baseDir}/scripts/generate_image_ref.py \
  --prompt "A woman walking in a forest" \
  --reference-image "https://example.com/character1.jpg" \
  --model "doubao-seedream-5-0-260128" \
  --aspect-ratio "16:9" \
  --token $NEODOMAIN_ACCESS_TOKEN

# Multiple reference images (up to 10)
python3 {baseDir}/scripts/generate_image_ref.py \
  --prompt "A conversation between two people" \
  --reference-image "https://example.com/character1.jpg" \
  --reference-image "https://example.com/character2.jpg" \
  --reference-image "https://example.com/character3.jpg" \
  --model "doubao-seedream-5-0-260128" \
  --aspect-ratio "16:9" \
  --token $NEODOMAIN_ACCESS_TOKEN

Parameter	Description	Default
--reference-image	Reference image URL(s) - can specify multiple (up to 10)	-

Video Generation

Get Available Models

python3 {baseDir}/scripts/video_models.py --token $NEODOMAIN_ACCESS_TOKEN

Generate Videos

Text-to-Video

python3 {baseDir}/scripts/generate_video.py \
  --prompt "A serene lake at dawn with mist rising from the water" \
  --model "veo3" \
  --generation-type "TEXT_TO_VIDEO" \
  --aspect-ratio "16:9" \
  --resolution "720p" \
  --duration "8s" \
  --token $NEODOMAIN_ACCESS_TOKEN

Image-to-Video

python3 {baseDir}/scripts/generate_video.py \
  --prompt "The camera slowly pans across the landscape" \
  --model "veo3" \
  --generation-type "IMAGE_TO_VIDEO" \
  --first-frame "https://example.com/image.jpg" \
  --aspect-ratio "16:9" \
  --resolution "720p" \
  --duration "8s" \
  --token $NEODOMAIN_ACCESS_TOKEN

Motion Control (Image + Video Reference)

python3 {baseDir}/scripts/motion_control.py \
  --image "https://example.com/ref_image.jpg" \
  --video "https://example.com/ref_video.mp4" \
  --prompt "Make the character dance" \
  --mode "pro" \
  --duration 5000 \
  --token $NEODOMAIN_ACCESS_TOKEN

Parameters

Parameter	Description	Default
--prompt	Text description for video	Required
--model	Model name: veo3, hailuo02, doubao	veo3
--generation-type	Type: TEXT_TO_VIDEO, IMAGE_TO_VIDEO, REFERENCE_TO_VIDEO	TEXT_TO_VIDEO
--first-frame	First frame image URL (for IMAGE_TO_VIDEO)	-
--last-frame	Last frame image URL (optional)	-
--image-urls	Reference images (comma-separated, for REFERENCE_TO_VIDEO)	-
--aspect-ratio	Video aspect: 16:9, 9:16, 1:1	16:9
--resolution	Resolution: 720p, 768p, 1080p	720p
--duration	Duration: 4s, 5s, 6s, 8s, 10s, 16s	8s
--fps	Frame rate	24
--seed	Random seed	Random
--generate-audio	Generate audio (true/false)	false
--enhance-prompt	Enhance prompt (true/false)	false
--output-dir	Where to save output	./output

Authentication

重要: 每次 token 过期或首次使用时，动态询问用户登录账号（手机号/邮箱），不要硬编码保存。

If you need to obtain an access token:

# 1. 发送验证码 (询问用户手机号或邮箱)
python3 {baseDir}/scripts/login.py --send-code --contact "用户手机号或邮箱"

# 2. 用户提供验证码后，登录获取 token
python3 {baseDir}/scripts/login.py --login --contact "用户手机号或邮箱" --code "验证码"

The login script will output an access token that you can store in NEODOMAIN_ACCESS_TOKEN.

Workflow

1. Authenticate (first time only): Get your access token
2. Get models: Check available models for your needs
3. Generate: Create content with appropriate parameters
4. Wait: Scripts automatically poll for results
5. Download: Images/videos saved to output directory

Output

• Images: *.jpg, *.png, or *.webp files
• Videos: *.mp4 files with thumbnail
• metadata.json with generation details