mcp-local-vision
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| VISION_NODE | No | Explicit node path for bin/launch.sh | |
| VISION_MODEL | No | Any vision-capable Ollama model | qwen3-vl:8b-instruct |
| VISION_PDF_DPI | No | Rasterization DPI for scanned pages | 200 |
| VISION_MAX_EDGE | No | Long-edge cap in px. Raise for dense small print | 1568 |
| VISION_MAX_PAGES | No | Cap on vision pages per call | 5 |
| VISION_TIMEOUT_MS | No | Per-model-request timeout | 300000 |
| VISION_OLLAMA_HOST | No | Ollama endpoint | http://127.0.0.1:11434 |
| VISION_ALLOWED_ROOTS | No | Colon-separated directory allowlist. Unset = unrestricted | |
| VISION_MAX_FILE_BYTES | No | Reject larger inputs | 209715200 |
| VISION_TEXT_PAGE_MIN_CHARS | No | Below this, a PDF page counts as scanned | 50 |
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 |
|---|---|
| read_imageA | Read a local image file (png, jpg, webp, gif, bmp, tiff, avif, heic) with a local vision model. Returns a verbatim transcription of all visible text plus a structural description of UI, charts, and diagrams. Use this for screenshots, error dialogs, design mockups, whiteboard photos, and graphs. Supply |
| read_pdfA | Read a local PDF. Pages with a text layer are extracted exactly and for free; scanned or image-only pages are rendered and read by a local vision model. Use |
| vision_doctorA | Report whether Ollama is reachable, whether the configured vision model is pulled, and whether poppler is installed for PDF rendering. Call this first when read_image or read_pdf fails. |
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 3 tools
Each tool targets a distinct input type: images, PDFs, and system diagnostics. The descriptions make boundaries clear, with no overlapping purposes.
Two tools follow the read_* pattern, but 'vision_doctor' deviates from the verb_noun style. Still, it's readable and predictable.
Three tools is well-scoped for a local vision server: two primary operations and one diagnostic helper. No redundancy or bloat.
Image and PDF reading are covered, along with setup diagnostics. Minor gaps like a standalone model list tool exist, but the core workflow is complete.