Optical Context MCP

Overview

Optical Context MCP is a Model Context Protocol (MCP) server: Compress OCR-heavy PDFs into dense packed images so agents can work with long visual documents. In practice that means any MCP-compatible AI assistant can call Optical Context MCP's tools directly — the model decides when to use them in a conversation or agent run.

It runs locally: the Python package optical-context-mcp (via uvx) speaks MCP over stdio on your machine. No credentials are required — it works out of the box. The current release is v0.1.4, published under the MIT license, with source at ChrBoebel/optical-context-mcp on GitHub.

Optical Context MCP is listed under File System on mcp.site and works with any MCP client — Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed and the rest of the ecosystem — using the install snippets below. If you maintain Optical Context MCP for ChrBoebel, claim this listing to verify ownership, earn the Verified badge, and keep the details current.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "optical_context_mcp": {
      "command": "uvx",
      "args": [
        "optical-context-mcp"
      ]
    }
  }
}