GenVR Skills

Generate images, videos, and process media using the GenVR API. Standalone Node.js CLI.

Security Scan

The skill's files, instructions, and requested environment variables are consistent with a CLI that talks to api.genvrresearch.com and only requires GENVR_API_KEY and GENVR_UID.

Purpose & Capability

Name/description, declared env vars, and the CLI code all match the stated purpose of interacting with the GenVR API (endpoints under api.genvrresearch.com). No unrelated credentials, binaries, or services are requested.

Instruction Scope

SKILL.md instructs the agent to run the Node CLI and to set GENVR_API_KEY and GENVR_UID only, which matches the code. One implementation detail is not documented: bin/genvr.js automatically loads a .env file from the current working directory if present, which could expose local environment values to the skill at runtime. The code only uses GENVR_API_KEY and GENVR_UID for API calls, but reading .env is implicit behavior that the SKILL.md does not call out.

Install Mechanism

There is no install spec (instruction-only), and the included code is pure Node.js with no external downloads or package installations. No high-risk install mechanism or external URL extraction is present.

Credentials

The skill declares only GENVR_API_KEY and GENVR_UID, which are appropriate and sufficient for the API usage. The implementation also loads a .env file from the working directory (if present) — this may read other local secrets though the code does not transmit arbitrary env values. Users should ensure their .env contains only intended GenVR credentials.

Persistence & Privilege

The skill does not request always:true or elevated privileges, does not modify other skills' configuration, and is not requesting persistent platform-level presence. Agent autonomous invocation remains allowed (platform default).

Assessment

This package appears to do what it claims: a small Node.js CLI that calls api.genvrresearch.com and needs GENVR_API_KEY and GENVR_UID. Before installing or running it, 1) verify the credentials come from the legitimate GenVR provider and use unique API keys (do not reuse high-privilege keys). 2) Remove or inspect any .env file in the working directory — the CLI will auto-load .env and may read any secrets stored there even though it only sends the GenVR API key to the service. 3) If you have concerns about provenance, run the CLI in an isolated environment (container or sandbox) and review the small bin/genvr.js source (it is bundled and readable) before use. 4) Note small documentation inconsistencies (index.js comment about Python and minor name/bin references) — these are likely editorial and not security issues but worth confirming the package source if you require a verified publisher.

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:

Env Var	Description
GENVR_API_KEY	Your GenVR API Token
GENVR_UID	Your GenVR User ID

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.