This paper reports on the implementation and evaluation of a Model Context Protocol (MCP) server for DraCor, enabling Large Language Models (LLM) to autonomously interact with the DraCor API. We conducted experiments focusing on tool selection and application by the LLM, employing a qualitative approach that includes systematic observation of prompts to understand how LLMs behave when using MCP tools, evaluating "Tool Correctness", "Tool-Calling Efficiency", and "Tool-Use Reliability". Our findings highlight the importance of "Docstring Engineering", defined as reflexively crafting tool documentation to optimize LLM-tool interaction. Our experiments demonstrate both the promise of agentic AI for research in Computational Literary Studies and the essential infrastructure development needs for reliable Digital Humanities infrastructures.
@article{arxiv.2508.13774,
title = {Agentic DraCor and the Art of Docstring Engineering: Evaluating MCP-empowered LLM Usage of the DraCor API},
author = {Peer Trilcke and Ingo Börner and Henny Sluyter-Gäthje and Daniil Skorinkin and Frank Fischer and Carsten Milling},
journal= {arXiv preprint arXiv:2508.13774},
year = {2025}
}
Comments
Preprint, submitted to the 2nd Workshop on Computational Drama Analysis at DraCor Summit 2025, September 03, 2025, Berlin, Germany