Skip to main content

MCP server that provides vision analysis tools via OpenAI-compatible API

Project description

vision-mcp

MCP server that provides vision analysis tools via an OpenAI-compatible API. Enables non-vision coding models to "see" images by routing image analysis to a separate vision-capable model.

Setup

Prerequisites

  • An OpenAI-compatible vision API endpoint (e.g., OpenAI, opencode.ai, or any provider supporting /v1/chat/completions with image_url content blocks)

Installation & Usage with uvx

The recommended way to run the server is via uvx — no manual installation needed:

uvx vision-mcp-server

Or install globally with uv:

uv tool install vision-mcp-server
vision-mcp-server

Environment Variables

Variable Required Default Description
VISION_API_BASE No https://api.openai.com/v1 Base URL for the OpenAI-compatible API
VISION_API_KEY Yes API key for authentication
VISION_MODEL No gpt-4o Model name to use for vision tasks

The server communicates over stdio using the MCP JSON-RPC protocol.

Tools

vision_analyze_image

Analyzes a single image file with an optional prompt.

| Parameter | Required | Description | |---|---|---|---| | image_path | Yes | Path to a local image file (PNG, JPEG, GIF, WebP) | | prompt | No | Guiding prompt (default: "Describe this image in detail.") |

vision_compare_images

Compares 2 to 8 image files simultaneously with a required prompt.

| Parameter | Required | Description | |---|---|---|---| | image_paths | Yes | List of 2-8 local image file paths | | prompt | Yes | Prompt describing what to compare |

Coding Agent Integration

opencode / Claude Code

Add to your project's .mcp.json:

{
  "mcpServers": {
    "vision": {
      "command": "uvx",
      "args": ["vision-mcp-server"],
      "env": {
        "VISION_API_BASE": "https://api.openai.com/v1",
        "VISION_API_KEY": "${VISION_API_KEY}",
        "VISION_MODEL": "gpt-4o"
      }
    }
  }
}

Note: The server installed via uvx/uv tool install creates the entry point vision-mcp-server (derived from the project name vision-mcp and script name server).

Supported Image Formats

  • PNG (.png)
  • JPEG (.jpg, .jpeg)
  • GIF (.gif)
  • WebP (.webp)

Maximum file size: 20 MB.

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

vision_mcp_server-0.1.0.tar.gz (4.3 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

vision_mcp_server-0.1.0-py3-none-any.whl (7.3 kB view details)

Uploaded Python 3

File details

Details for the file vision_mcp_server-0.1.0.tar.gz.

File metadata

  • Download URL: vision_mcp_server-0.1.0.tar.gz
  • Upload date:
  • Size: 4.3 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.9 {"installer":{"name":"uv","version":"0.10.9","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for vision_mcp_server-0.1.0.tar.gz
Algorithm Hash digest
SHA256 1eca5486f703cd6458e51de85fe37b1c984ecc725480b8c65187c4ae790a9a4e
MD5 04183ecb1a2603d0649bacad1935555a
BLAKE2b-256 15898035afc554ac048a8a20254d3c5cf3ec9f2218fb48ff535ca0789160e5f9

See more details on using hashes here.

File details

Details for the file vision_mcp_server-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: vision_mcp_server-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 7.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.9 {"installer":{"name":"uv","version":"0.10.9","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for vision_mcp_server-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 84a133d8394f7b0339a0539862d5d08485884dd1d53df3a1f09ee96502e63387
MD5 a55827ba8c16dc49464db40cc54acc56
BLAKE2b-256 6b7a42f676da161f01de119710dfef52bce5f44a24c366ba80b9171fb454594b

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page