LLMWhisperer MCP server

LLMWhisperer has support for MCP server which can be used to integrate the functionality of LLMWhisperer with a MCP host.

Features

Processes documents (PDF, DOCX, XLSX, CSV, images) using the LLMWhisperer API to extract layout-preserved text, supporting multiple processing modes and output formats.

Tools

extract_text - Submits a file to the LLMWhisperer API, polls for processing, and retrieves the extracted text. Supports advanced options for page selection, line splitting, and output formatting.

Configuration

Getting an API Key

To use the tool, one needs to set the LLMWHISPERER_API_KEY environment variable. For obtaining the key, please check the details under Getting Started.

Usage with Claude Desktop

One can integrate the LLMWhisperer tool with Claude Desktop by adding the following configuration to the claude_desktop_config.json replacing the values of LLMWHISPERER_API_KEY with the right values. Also, since the tool will need a file to be processed, the file needs to be mounted in the container. The sample below uses /tmp as the mount point. This can be changed to any other location.

Docker

{
  "mcpServers": {
        "extract_text": {
            "command": "docker",
            "args": [
                "run",
                "-i",
                "--rm",
                "-v",
                "/tmp:/tmp",
                "-e",
                "LLMWHISPERER_API_KEY",
                "unstract/mcp-server",
                "llm_whisperer"
            ],
            "env": {
                    "LLMWHISPERER_API_KEY": ""
            }
        }
    }
}

Once the configuration is added, user can use extract_text as a tool to extract text from the supported documents.

Sample prompt

Extract text from the document /tmp/sample-bank_statement.pdf

This should invoke the tool and get the extracted text as the response.