Vision MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GEMINI_API_KEY | No | Fallback Gemini API key used when VISION_MCP_GEMINI_API_KEY is not set. | |
| VISION_MCP_CONFIG | No | Path to the configuration file. | |
| VISION_MCP_TIMEOUT | No | Runtime timeout in seconds. Defaults to 120. | 120 |
| VISION_MCP_WORKDIR | No | Runtime working directory. If null, a temporary directory is used per task. | |
| VISION_MCP_AGY_MODEL | No | Model to use for the AGY provider. If null, the provider default is used. | |
| VISION_MCP_LOG_LEVEL | No | Log level: ERROR, WARNING, INFO, or DEBUG. Defaults to INFO. | INFO |
| VISION_MCP_AGY_COMMAND | No | Executable for the AGY provider. Defaults to 'agy'. | agy |
| VISION_MCP_CODEX_MODEL | No | Model to use for the Codex provider. If null, the provider default is used. | |
| VISION_MCP_GEMINI_MODEL | No | Model to use for the Gemini provider. If null, the provider default is used. | |
| VISION_MCP_CODEX_COMMAND | No | Executable for the Codex provider. Defaults to 'codex'. | codex |
| VISION_MCP_GEMINI_API_KEY | No | Gemini API key. This takes precedence over GEMINI_API_KEY. | |
| VISION_MCP_OPENCODE_MODEL | No | Model to use for the OpenCode provider. If null, the provider default is used. | |
| VISION_MCP_MAX_CONCURRENCY | No | Maximum number of concurrent tasks. Defaults to 2. | 2 |
| VISION_MCP_OPENCODE_COMMAND | No | Executable for the OpenCode provider. Defaults to 'opencode'. | opencode |
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 |
|---|---|
| ui_to_artifactB | Convert a UI screenshot into an artifact (code, prompt, spec or description). |
| extract_text_from_screenshotC | Extract visible text (OCR), source code, terminal/config content verbatim. |
| diagnose_error_screenshotB | Diagnose an error/stack trace shown in a screenshot. |
| understand_technical_diagramC | Understand an architecture/flowchart/UML/ER/system diagram. |
| analyze_data_visualizationB | Analyze a chart/plot (trends, anomalies, comparisons, distribution). |
| ui_diff_checkB | Compare EXPECTED (first) vs ACTUAL (second) UI for visual regression. |
| analyze_imageC | General visual analysis of an image. |
| analyze_videoD | Analyze a video (mp4/mov/m4v). |
| image_analysisD | Alias for analyze_image. |
| video_analysisD | Alias for analyze_video. |
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 10 tools
Several tools have direct aliases (analyze_image/image_analysis, analyze_video/video_analysis) causing ambiguity. The other image-analysis tools (data visualization, UI diff, error screenshot, diagram understanding) overlap in scope but have distinct purposes, so some differentiation is possible.
Most tools follow a verb_noun pattern (analyze_data_visualization, analyze_image, extract_text_from_screenshot), but several deviate (ui_diff_check, ui_to_artifact, image_analysis, video_analysis). The mixed conventions and presence of aliases make naming inconsistent.
Ten tools is within a reasonable range for a vision server, but two are redundant aliases, effectively reducing the count to eight. The set is not bloated, but the aliases needlessly inflate the number.
The server covers a broad range of vision tasks: general image/video analysis, chart understanding, UI regression, OCR, error diagnosis, diagram comprehension, and UI conversion. Missing some common vision features like object detection or face analysis, but for its UI/development focus, the coverage is solid.