FastMCP FastMCP
Sign In
A

AI Vision MCP Server

Enables AI-powered image and video analysis using Google Gemini and Vertex AI models. Supports analy...

58 views
0 installs
Updated Feb 5, 2026
Not audited
Cloud Platforms AI and Machine Learning
Tools I Recommend
ClaudeKit ClaudeKit Sponsor
Enables AI-powered image and video analysis using Google Gemini and Vertex AI models. Supports analyzing single or multiple images, detecting objects with bounding boxes, and video content analysis through natural language prompts.

  1. Open the FastMCP connection interface (click “Install Now”) and prepare to fill the environment fields there.

    • You will enter the environment variable names/values directly into the FastMCP connection form (the same keys shown in the README). (github.com)

  2. If you’re using the Google AI Studio / Gemini provider (recommended)

    1. Set IMAGE_PROVIDER and VIDEO_PROVIDER to "google" in the FastMCP form. (github.com)
    2. Obtain a Gemini API key:
      • Sign in to Google AI Studio (aistudio.google.com).
      • Open Projects (or Import Projects if your Google Cloud project is not yet listed), then go to the API keys / “Get API key” area.
      • Create a new API key and copy it (the key is shown only once). (ai.google.dev)
    3. In the FastMCP form fill:
      • IMAGE_PROVIDER = google
      • VIDEO_PROVIDER = google
      • GEMINI_API_KEY =
    4. Keep the API key private (do not commit to source control). (ai.google.dev)

  3. If you’re using the Vertex AI provider (production / alternative)

    1. Set IMAGE_PROVIDER and VIDEO_PROVIDER to "vertex_ai" in the FastMCP form. (github.com)
    2. Create or choose a Google Cloud project, enable Vertex AI and Cloud Storage APIs, and ensure billing is enabled for that project. (docs.cloud.google.com)
    3. Create a service account and download a JSON key:
      • In the Google Cloud Console go to IAM & Admin → Service Accounts → Create Service Account.
      • Grant needed roles (e.g., Vertex AI User / roles/aiplatform.user and Storage Admin or appropriate storage permissions), finish, then open that service account → Keys → Add Key → Create new key → JSON → Download. You will get a service-account JSON file (keep it secure). (docs.cloud.google.com)
    4. Create (or pick) a Google Cloud Storage bucket for model/video/object storage: create a globally-unique bucket name in Cloud Storage → Create. Note the exact bucket name (it must be globally unique). (cloud.google.com)
    5. In the FastMCP form fill:
      • IMAGE_PROVIDER = vertex_ai
      • VIDEO_PROVIDER = vertex_ai
      • VERTEX_CREDENTIALS = /absolute/path/to/your-service-account.json (upload or point to the local path the MCP host will have access to)
      • GCS_BUCKET_NAME = your-gcs-bucket-name

  4. After filling values in FastMCP, save/confirm the connection and start the MCP server

    • Use the client’s “Install Now” / save button to apply the envs and start the MCP integration. The MCP will read GEMINI_API_KEY or VERTEX_CREDENTIALS + GCS_BUCKET_NAME at startup. (github.com)

  5. Quick checks and security reminders

    1. Verify the MCP startup logs for authentication success / any credential errors.
    2. Never commit API keys or service-account JSON to git; restrict access to the JSON file and rotate/delete keys if exposed. (ai.google.dev)

If you want, tell me which provider you’ll use (google or vertex_ai) and I’ll produce the exact field values and a short copy-paste checklist for the FastMCP “Install Now” form.

Quick Start

View on GitHub

More for Cloud Platforms

View All →

More for AI and Machine Learning

View All →

Similar MCP Servers

A

AI Vision

Integrates with Google's Gemini and Vertex AI models to analyze images, compare multiple images, and process video content with intelligent file handling that automatically optimizes upload strategies for different file sizes.

AI and Machine Learning
355
2
G

Google AI Studio

Integrates with Google AI Studio/Gemini API to process multimodal content including images, videos, audio, PDFs, and text files for content generation, analysis, and document conversion tasks.

AI and Machine Learning
626
1
U

Universal Image Generator

Provides multi-provider image generation and transformation capabilities across Google Gemini, ZhipuAI, and Alibaba Bailian with automatic prompt translation and optimization for each provider's preferred language, supporting URL-based editing with mask support and flexible input methods including base64 encoding, file paths, and public URLs.

AI and Machine Learning
451
0

Report Issue

Thank you! Your issue report has been submitted successfully.