MetaRouter Image MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| METAROUTER_API_KEY | Yes | 必填，MetaRouter API Key | |
| METAROUTER_TIMEOUT | No | 请求超时秒数，最小 30 秒 | 300 |
| METAROUTER_SAVE_DIR | No | 默认输出目录，必须位于根目录内 | |
| METAROUTER_INPUT_ROOT | No | 可选；设置后只允许读取该目录内的参考图片 | |
| METAROUTER_IMAGE_MODEL | No | 常规生图模型 | gpt-image-2 |
| METAROUTER_MAX_INPUT_MB | No | 单张本地输入图片的大小上限 | 10 |
| METAROUTER_SAVE_DIR_ROOT | No | MCP 允许写入的根目录 | ~/Pictures/MetaRouter |
| METAROUTER_IMAGE_PRO_MODEL | No | 大尺寸自动使用的模型 | gpt-image-2-pro |
| METAROUTER_TRUSTED_IMAGE_HOSTS | No | 允许下载结果图片的 HTTPS 域名，逗号分隔 | metarouters.org |
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| image_generateA | 通过 MetaRouter 文生图。 Args: prompt: 图片描述，建议包含主体、构图、风格、光线以及禁止内容。 size: WxH，例如 1024x1024、1536x1024、2048x2048。宽高须为 8 的倍数。 model: 可选模型；留空默认 gpt-image-2，超大尺寸自动使用 gpt-image-2-pro。 n: 生成数量，1 到 4；高分辨率建议使用 1。 quality: 可选质量参数，留空使用后端默认。 output_dir: 可选保存目录，必须位于配置的安全输出根目录下。 |
| image_editA | 使用一张本地参考图，通过 MetaRouter 改图或图生图。 Args: image_path: 本地 PNG、JPEG 或 WebP 文件的绝对路径。 prompt: 编辑要求，应明确哪些内容保留、删除或修改。 size: 输出尺寸 WxH，宽高须为 8 的倍数。 model: 可选模型；留空自动选择。 quality: 可选质量参数。 output_dir: 可选保存目录，必须位于安全输出根目录下。 |
| image_batch_editA | 对多张本地图片逐张执行同一编辑指令，适合批量换背景或统一风格。 image_paths 最多 20 张；每张图片会产生一次独立的计费请求。 |
| image_multi_referenceA | 融合 2 到 10 张本地参考图，生成一张新图。 prompt 中应写明每张参考图分别提供哪些构图、配色、人物或产品特征。 |
| server_infoA | 返回当前 MetaRouter MCP 的安全运行配置，不返回完整 API Key。 |
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 5 tools
Each image tool has a distinct purpose: generation, single edit, batch edit, and multi-reference fusion. server_info is clearly separate as configuration info. No two tools are easily confused.
The image_ prefix unifies the image tools, but the suffixes mix verb forms (generate, edit) with noun phrases (multi_reference). server_info deviates from the prefix but is a typical meta-tool.
With five tools, the server is well-scoped, covering the essential image generation and editing workflows without redundancy. Each tool has a clear role.
The toolset covers text-to-image, single-image editing, batch editing, and multi-reference fusion, which are the major image operations. server_info provides configuration context, leaving no obvious gaps.