Nano Banana Veo

Generate images with Nano Banana (Gemini 3 Pro Image) and animate them into videos with Veo 3.1. Use when creating AI-generated visual assets for websites, l...

Security Scan

The skill's code and instructions match its stated purpose (generating images with Gemini and animating with Veo); the main inconsistency is that the registry metadata does not list the required GEMINI_API_KEY even though the SKILL.md and script require it.

Purpose & Capability

Name/description (Nano Banana + Veo) align with the included script and SKILL.md: the script calls Gemini image models and the Veo endpoints. However, the registry metadata lists no required environment variables while SKILL.md and the script explicitly require GEMINI_API_KEY — a metadata omission/inconsistency.

Instruction Scope

SKILL.md instructs installing google-genai and requests, running the provided script with prompts and output paths. The runtime instructions and examples match what the script actually does (generate image, encode it, call Veo API, poll for results). The instructions do not request unrelated files or credentials beyond GEMINI_API_KEY.

Install Mechanism

This is an instruction-only skill with one included script; there is no install spec downloading arbitrary code or running external installers. Dependencies are standard Python packages (google-genai, requests) installed via pip as documented.

Credentials

The script requires a single API key (GEMINI_API_KEY) which is proportionate to calling Google generative APIs. The inconsistency: registry metadata/manifest did not declare this required env var or a primary credential, while both SKILL.md and scripts require it. No other secrets or unrelated credentials are requested.

Persistence & Privilege

The skill is not always-enabled and does not request elevated platform privileges. It does not modify other skills or system-wide configuration. Autonomous invocation is allowed by default (normal) but not combined with other red flags.

Assessment

This skill appears to do what it claims: it generates an image via Gemini and then calls Veo to produce a short video. Before installing: 1) Note that the registry metadata omitted GEMINI_API_KEY — you must set GEMINI_API_KEY in your environment for it to work. 2) Treat that API key as sensitive (it grants access to your Google generative API quota/billing) and only provide it if you trust the skill/source. 3) Review and run the script in an isolated environment (virtualenv or container) if you want to limit exposure. 4) If you plan to use this in production, ask the publisher to update the registry metadata to declare GEMINI_API_KEY as the primary credential so the manifest accurately reflects required secrets.

Nano Banana + Veo Workflow

Generate premium images and animate them into cinematic videos using Google's Gemini API.

Quick Start

# Generate just an image
uv run {baseDir}/scripts/generate.py \
  --image-prompt "iPhone floating against dark background, premium product photography" \
  --output-image hero.png

# Generate image + video
uv run {baseDir}/scripts/generate.py \
  --image-prompt "iPhone floating against dark background, premium product photography" \
  --video-prompt "iPhone gently rotating and floating, smooth seamless motion, cinematic lighting" \
  --output-image hero.png \
  --output-video hero.mp4 \
  --video-duration 4

Parameters

Parameter	Required	Default	Description
--image-prompt	✅	-	Prompt for image generation
--video-prompt	❌	image-prompt	Prompt for video animation
--output-image	✅	-	Output path for image (.png/.jpg)
--output-video	❌	-	Output path for video (.mp4)
--resolution	❌	1K	Image resolution: 1K, 2K, 4K
--video-duration	❌	4	Video seconds (4-8, Veo requirement)
--video-resolution	❌	720p	Video quality: 720p, 1080p, 4k

Prerequisites

• GEMINI_API_KEY environment variable set
• google-genai and requests Python packages

Installation

# Install dependencies
pip install google-genai requests

# Or with uv
uv pip install google-genai requests

Workflow Examples

Hero Section Asset

uv run {baseDir}/scripts/generate.py \
  --image-prompt "Sleek iPhone 15 Pro showing Islamic prayer app interface, dark green and gold accents, floating against pure black background, dramatic studio lighting, premium product photography, 8K quality" \
  --video-prompt "iPhone floating and gently rotating in space, subtle purple glow, smooth seamless boomerang motion, premium cinematic lighting, dark background" \
  --output-image namazlock-hero.png \
  --output-video namazlock-hero.mp4 \
  --video-duration 4

Feature Showcase

uv run {baseDir}/scripts/generate.py \
  --image-prompt "Minimalist smartphone displaying prayer time interface, clean UI, dark mode with gold accents, studio lighting" \
  --output-image feature.png \
  --resolution 2K

Output

• Images are saved in PNG format
• Videos are saved in MP4 format (H.264 encoded)
• Veo generation takes ~30-60 seconds (async polling handled automatically)

Notes

• Veo 3.1 requires 4-8 second duration (enforced by API)
• Video generation is asynchronous — script handles polling automatically
• Use {baseDir} placeholder in paths — it resolves to the skill directory