genvr-skills（genvr 技能）

GenVR Skills — AI Media Generation & Utilities

Generate images, videos, and perform various media processing tasks using the GenVR API. This standalone Node.js toolkit provides a pure CLI for interacting with GenVR's asynchronous task pipeline.

IMPORTANT: Most operations require a GENVR_API_KEY and a GENVR_UID. Get them from api.genvrresearch.com.

Usage

Run via npx directly from the package:

List available models:

npx genvr-skills list

Generate content (unified):

npx genvr-skills generate --category imagegen --subcategory google_imagen3 prompt="A futuristic city in VR" height=1024 width=1024

Edit an image with Nano Banana 2:

npx genvr-skills generate --category imgutils --subcategory google_nano_banana_2 image_url="https://media.genvr.com/assets/..." prompt="add a futuristic helmet"

Check job status:

npx genvr-skills status --job-id "TASK_ID" --category "imagegen" --subcategory "google_imagen3"

Authentication

Authentication is handled via environment variables:

Core Commands

list

Lists all available categories and subcategories.

• --json: Output as raw JSON.

generate

The primary command for starting new generation tasks.

• --category: The category of the model (e.g., imagegen).
• --subcategory: The specific model variant (e.g., google_imagen3).
• --filename: Optional output filename.
• --no-wait: Return immediately after starting the job.
• key=value: Model parameters (e.g., prompt="cat", steps=25).

status

Retrieves the status or result of an existing task.

• --job-id: The ID returned by generate.
• --category: Required for v2 status check.
• --subcategory: Required for v2 status check.

Examples

Generating a Cinematic 360 Panorama

npx genvr-skills generate --category 3dgen --subcategory panorama prompt="A hyper-realistic cinematic 360 panorama of an ancient bioluminescent forest at night, 8k resolution, ethereal lighting"

Professional Video Generation (Image-to-Video)

Use a high-quality model like Google Veo 3 for image-to-video translation:

npx genvr-skills generate --category videogen --subcategory google_veo3_i2v image_url="https://media.genvr.com/assets/..." prompt="The camera cinematicly glides through the scene as the bioluminescent plants begin to glow intensely, soft wind blowing through the leaves, masterpiece quality"

Filename Pattern

Generated files follow the pattern: yyyy-mm-dd-hh-mm-ss-category.ext.

Tips

• Run list frequently to see new tools added to the platform.
• Use --no-wait for video generation tasks, as they may take several minutes.
• Most models follow the standard OpenAPI schema; details at api.genvrresearch.com.