Mcp Google Vertex — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Mcp Google Vertex (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
A Model Context Protocol (MCP) server that provides AI-powered image and video generation capabilities using Google Vertex AI's Imagen and Veo models.
Generate AI images using the configured Imagen model (set via VERTEX_AI_IMAGE_MODEL).
Parameters:
| Parameter | Type | Default | Description | ||||
|---|---|---|---|---|---|---|---|
prompt | string | required | Text description of the image to generate | ||||
numberOfImages | number (1-8) | 1 | Number of images to generate | ||||
aspectRatio | 1:1 \ | 3:4 \ | 4:3 \ | 9:16 \ | 16:9 | 1:1 | Aspect ratio |
imageSize | 1K \ | 2K | 2K | Output resolution | |||
outputMimeType | image/png \ | image/jpeg | image/png | Output format | |||
negativePrompt | string | — | Things to avoid in the image | ||||
guidanceScale | number (1-20) | — | How closely the model follows the prompt | ||||
seed | number | — | Random seed for reproducible results | ||||
enhancePrompt | boolean | false | Auto-enhance the prompt before generation |
Example:
{
"name": "generate-image",
"arguments": {
"prompt": "A serene mountain landscape at sunset with a lake",
"aspectRatio": "16:9",
"numberOfImages": 2
}
}Generate AI videos using the configured Veo model (set via VERTEX_AI_VIDEO_MODEL).
Parameters:
| Parameter | Type | Default | Description | ||
|---|---|---|---|---|---|
prompt | string | required | Text description of the video to generate | ||
numberOfVideos | number (1-4) | 1 | Number of videos to generate | ||
durationSeconds | number (4-8) | 8 | Clip length in seconds (4, 6, or 8) | ||
aspectRatio | 16:9 \ | 9:16 | 16:9 | Aspect ratio | |
resolution | 720p \ | 1080p \ | 4K | 1080p | Video resolution |
seed | number | — | Random seed for reproducible results | ||
negativePrompt | string | — | Things to avoid in the video | ||
enhancePrompt | boolean | true | Auto-enhance the prompt before generation | ||
generateAudio | boolean | false | Generate audio alongside the video | ||
lastFrame | string | — | Image to use as the last frame (image-to-video) | ||
referenceImages | array | — | Reference images to guide generation (see below) |
Reference images (provide either a local file path, Cloud Storage URI, or public URL):
/path/to/image.pnggs://my-bucket/image.jpghttps://cdn.example.com/image.jpgSupported formats: JPEG, PNG. Maximum size: 10 MB.
referenceImages supports up to 3 ASSET images or 1 STYLE image.
Example — text to video:
{
"name": "generate-video",
"arguments": {
"prompt": "A butterfly flying through a garden of flowers",
"durationSeconds": 8,
"aspectRatio": "16:9",
"resolution": "1080p"
}
}Example — image reference:
{
"name": "generate-video",
"arguments": {
"prompt": "The product spinning on a white background",
"referenceImages": [
{
"image": "/path/to/product.png",
"referenceType": "ASSET"
}
]
}
}Add to your claude_desktop_config.json:
{
"mcpServers": {
"google-vertex": {
"command": "npx",
"args": ["mcp-remote", "http://localhost:3005/mcp"]
}
}
}Add to your .vscode/mcp.json:
{
"servers": {
"google-vertex": {
"type": "http",
"url": "http://localhost:3005/mcp"
}
}
}Test your server with the MCP Inspector:
npx @modelcontextprotocol/inspectorThen connect to: http://localhost:3005/mcp
The server follows clean architecture principles with separation of concerns:
The server includes comprehensive error handling:
MIT
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.