Image Gen Pro MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GEMINI_API_KEY | No | Your AI Studio key. GOOGLE_API_KEY works too, since Google's own tools read that one | |
| GOOGLE_API_KEY | No | Alias for GEMINI_API_KEY; Google's tools read this one too | |
| OPENAI_API_KEY | No | Your OpenAI key. Reaches GPT Image 2 directly | |
| AI_GATEWAY_API_KEY | No | Vercel AI Gateway. Reaches everything, and is the only route to Recraft, Seedream, Grok Imagine and Arrow | |
| IMAGE_GEN_OUTPUT_DIR | No | Where images are written. Point it at your project's image folder and the paths come back ready to use | |
| IMAGE_GEN_TIMEOUT_MS | No | How long to wait. Image models are slow; three minutes is not generous | 180000 |
| IMAGE_GEN_MAX_PER_CALL | No | Ceiling on one request, whatever the assistant asks for | 4 |
| IMAGE_GEN_MAX_PER_HOUR | No | Images per rolling hour. 0 removes the cap, knowing what that means for your bill | 60 |
| IMAGE_GEN_PREFER_GATEWAY | No | Route through the gateway even when a direct key exists. Turn on to put all the billing in one place | false |
| IMAGE_GEN_DEFAULT_PROVIDER | No | Which vendor runs when the assistant names none. Set to 'google' if your OpenAI account can't reach GPT Image 2 | openai |
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 | {} |
| logging | {} |
| completions | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| generate_imageA | Generate one or more images from a description and write them to disk, returning their paths. Optionally takes REFERENCE IMAGES (paths on this machine) and extra text context, both of which steer the result: use them to match an existing style, keep a character or product consistent, or edit an image you already have. Defaults to OpenAI GPT Image 2; pass |
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 1 tool
Only one tool exists, so there is no possibility of confusing it with other tools. The tool's description clearly delineates its multiple functions (generation, editing, style selection, SVG), but since it is the sole tool, disambiguation is trivially perfect.
The single tool is named 'generate_image', which follows a clear verb_noun pattern. With only one tool, there is no inconsistency or mixed conventions to evaluate, making the naming fully consistent.
The server has just one tool, which is on the lower end of the typical range. However, the tool is comprehensive, covering generation, editing, and multiple model options, so it justifies its presence. A slightly higher count might be expected, but the single tool is not trivial.
For an image generation server, the tool covers core operations: generating from text, editing with reference images, and producing SVG output. Minor gaps exist (e.g., no listing or deletion of generated images), but these are non-essential and do not block typical workflows.