Efax

You run an MCP server that converts OpenText Fax Server documents (eFax) into structured JSON, supporting PDF, TIFF, and CCD XML formats with OCR and metadata extraction to enable seamless downstream processing and data integration.

How to use

To use the eFax to JSON MCP Server, configure your MCP client to connect to the local server process or to a remote MCP endpoint, then issue the conversion commands to transform incoming eFax documents into standardized JSON. You can convert a single document, batch multiple documents, validate the resulting JSON, or inspect file information and supported formats. The available workflow patterns include:

• Convert a single document with optional OCR and metadata extraction
• Batch convert all documents in a directory with optional error handling
• Validate the produced JSON against the expected structure
• Retrieve file information for input documents
• List supported input formats (PDF, TIFF, CCD XML) for planning workflows These patterns help you integrate eFax data into your data lake, CMS, or analytics pipelines with consistent output.

How to install

# Prerequisites
node v18+
system-level Tesseract OCR (required for OCR on images)
  - Ubuntu/Debian: sudo apt-get install tesseract-ocr
  - macOS: brew install tesseract
  - Windows: Download from UB Mannheim releases

# 1. Create project directory
mkdir efax-mcp-server
cd efax-mcp-server

# 2. Initialize and install dependencies
npm init -y
npm install @modelcontextprotocol/sdk pdf-parse sharp tesseract.js xml2js
npm install -D @types/node @types/pdf-parse @types/xml2js typescript ts-node

# 3. Create directory structure
mkdir -p src/{types,processors,utils}
mkdir -p tests/test-files
mkdir -p docs

# 4. Add source files
# Paste the provided source code into respective files

# 5. Build the project
npm run build

Configuration and usage notes

Configure your MCP client to connect to the eFax converter as an MCP server. The following example shows a local stdio configuration that runs the server via Node using the built distribution.

{
  "mcpServers": {
    "efax-converter": {
      "command": "node",
      "args": ["/path/to/efax-mcp-server/dist/server.js"]
    }
  }
}

Tools you can use with the server

You can perform common operations to convert documents, batch process directories, validate outputs, and inspect files or formats. The available commands include:

• Convert Single Document
• Batch Convert Documents
• Validate JSON Output
• Get File Information
• List Supported Formats These tools enable end-to-end ingestion of eFax data into your JSON-based workflows.

Security and maintenance notes

Ensure the MCP server runs in a secured environment and that access to input documents is controlled. Monitor OCR workloads, as they are CPU-intensive, and consider running OCR as a constrained batch to manage resource usage. Regularly update Tesseract data packs and keep dependencies up to date to minimize vulnerabilities.

Troubleshooting

If OCR seems not to run, verify the Tesseract installation and language data availability. Check that input files are accessible and that there is enough memory for large TIFFs. For permission issues, confirm read access to input files and write access to output locations, and ensure the MCP server process has the necessary filesystem permissions.

Available tools

convert_single_document

Converts a single eFax document to JSON with optional OCR and metadata extraction.

batch_convert_documents

Converts all documents in a directory to JSON in batch mode, with configurable output and error handling.

validate_json_output

Validates the produced JSON against the expected eFax JSON structure.

get_file_info

Retrieves information about a specific input document, such as path, size, and pages.

list_supported_formats

Reports supported input formats (PDF, TIFF, CCD XML) for the MCP server.