mcp-multivision-server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MCP_HOST | No | SSE listen host | 0.0.0.0 |
| MCP_PORT | No | SSE listen port | 8093 |
| MCP_TRANSPORT | No | Transport type: stdio or sse | stdio |
| MCP_FFMPEG_BIN | No | Path to ffmpeg binary (default: from PATH) | |
| MCP_FFPROBE_BIN | No | Path to ffprobe binary (default: from PATH) | |
| MCP_VISION_MODEL | No | Vision model name (required for image/video analysis) | |
| MCP_VISION_API_KEY | No | Vision API key (required for image/video analysis) | |
| MCP_VISION_TIMEOUT | No | Request timeout in seconds | 60 |
| MCP_VISION_BASE_URL | No | Vision API base URL (required for image/video analysis) | |
| MCP_VISION_PROVIDER | No | Vision provider name (currently openai compatible) | openai |
| MCP_VISION_WORK_DIR | No | Temporary directory for frames/audio | /tmp/mcp-vision |
| MCP_VISION_CACHE_DIR | No | Cache directory | /tmp/mcp-vision-cache |
| MCP_VISION_MAX_FRAMES | No | Max frames sent to model in analyze_video | 8 |
| MCP_VISION_MAX_TOKENS | No | Max tokens for generation | 1024 |
| MCP_VISION_MAX_RETRIES | No | Max retries on failure | 3 |
| MCP_VISION_TEMPERATURE | No | Sampling temperature | 0.2 |
| MCP_VISION_CACHE_ENABLED | No | Enable cache for analysis results | false |
| MCP_VISION_MAX_IMAGE_SIZE | No | Max image size in bytes (20MB) | 20971520 |
| MCP_VISION_ALLOWED_IMAGE_FORMATS | No | Allowed image formats | jpeg,png,webp,gif,bmp,tiff |
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
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| vision_analyze_imageA | 图片理解问答：调用云端视觉大模型对图片进行描述、问答、OCR、图表/UI/报错分析等。可用 prompt 自由提问，或用 preset 选择任务类型。 |
| vision_analyze_videoA | 视频理解：将视频原生交给云端多模态大模型（如 Qwen3.5，通过 DashScope video_url）进行理解——抽帧与时序对齐由模型服务端完成，本地不做任何处理。推荐传视频的 http(s) URL；本地文件会转 base64（受大小限制）。 |
| vision_image_metadataA | 图片元信息：读取尺寸、格式、颜色模式与 EXIF（含拍摄时间、GPS 若有），本地 Pillow 解析，不调用云端。 |
| vision_get_server_statusA | 查询服务状态：云端视觉模型是否已配置、当前模型与配置项。 |
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 4 tools
Each tool targets a distinct function: image analysis, video analysis, server status, and metadata extraction. No overlap in purpose, and descriptions clearly differentiate them.
Tools follow a consistent 'vision_<verb>_<target>' pattern for three tools, but 'vision_image_metadata' lacks a verb, though it remains readable.
Four tools is well-scoped for a vision server, covering analysis, status, and metadata without unnecessary proliferation.
The tool surface covers core operations: image/video analysis, server status, and image metadata. No obvious gaps for the stated purpose.