Integrates with Google's Agent Development Kit (ADK) to enable LLM agents to interact with Aucterra's document understanding services.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Aucterra MCP ServerExtract the invoice number and total amount from this PDF document."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Aucterra MCP Server
This repository configures an MCP-compatible server for Aucterra's Document Understanding APIs using the aucterra-mcp package.
It enables LLM agents to interact with Aucterra's document classification, extraction, etc. services using Google's Agent Development Kit (ADK).
🔧 Configuration
Include this block in your mcpServers configuration (e.g., config.json or mcp.yaml):
🔐 Required Environment Variables
Variable | Description |
| Your API key for accessing Aucterra's services |
Related MCP server: MCP LLMS-TXT Documentation Server
⚙️ Tool Behavior
This MCP tool provides structured access to Aucterra's:
📁 Document Classification
🗂️ Key-Value Field Extraction (Simple + List fields)
The tool accepts pdf or image files and returns structured JSON output.
✅ Agent Integration (Google ADK)
To use this tool within your LlmAgent, configure it as follows:
Add this tool to your agent via tools=[aucterra_tool].
📚 Example Use Cases
The agent will use the Aucterra MCP server to extract fields, filling in missing values (e.g., using the same value for field_key and field_name if only one is provided).
📦 Dependencies
Ensure pipx is installed and available on your system.
Install it via:
🛠️ Issues
If you encounter issues, ensure your API key is correct and the tool is up to date: