gemini-image-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LOG_LEVEL | No | Log level (debug, info, or error) | info |
| OUTPUT_DIR | No | Default directory for saved images | ~/gemini-images |
| DEFAULT_MODEL | No | Default Gemini model | gemini-2.5-flash-image |
| GEMINI_API_KEY | Yes | Your Gemini API key | |
| MAX_COST_PER_HOUR | No | Max estimated cost (USD) per rolling hour (0 = unlimited) | 0 |
| REQUEST_TIMEOUT_MS | No | API request timeout in milliseconds | 60000 |
| SESSION_TIMEOUT_MS | No | Multi-turn session expiry in milliseconds | 1800000 |
| MAX_REQUESTS_PER_HOUR | No | Max image generations per rolling hour (0 = unlimited) | 0 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| generate_imageA | Generate or edit images using Google Gemini. Provide just a prompt for text-to-image generation. Add image file paths to edit or use reference images. Set removeBackground to get a transparent PNG cutout in one call (local AI matte; works on any subject, no extra API cost). Returns the saved file path, model used, token counts, and estimated cost. Advanced inputs (video-to-image, thinking depth, image-search grounding): see the README's Advanced Features section. |
| process_imageA | Process an existing image locally using sharp. Crop, resize, remove background, convert format, or trim whitespace. Free, fast, no API calls. For AI-powered editing (style changes, complex background removal), use generate_image with the image as input instead. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 2 tools
The two tools are largely distinct: one performs AI-powered generation/editing via Gemini, the other does local sharp-based processing. However, both support background removal, which could cause slight confusion, though the descriptions clarify the difference in approach and cost.
Both tool names follow a consistent verb_noun pattern (generate_image, process_image) using snake_case. The naming is clear, predictable, and parallel.
With only two tools, the server feels thin for a general-purpose image MCP. While they cover both AI generation/editing and local processing, the scope might benefit from additional tools like image analysis or format conversion, but this is borderline.
The core workflows of generating images from text, editing with reference images, and local image processing are covered. Minor gaps exist, such as no explicit tool for image analysis or advanced compositing, but the two tools together handle most primary use cases.