Doc2Agent: Scalable Generation of Tool-Using Agents from API Documentation
This addresses the problem of creating agents for arbitrary domains with complex APIs, offering a generalizable solution, though it appears incremental as it builds on existing tool-generation methods.
The paper tackles the challenge of building tool-using agents from unstructured API documentation by proposing Doc2Agent, a scalable pipeline that generates executable tools and achieves a 55% relative performance improvement with 90% lower cost on the WebArena benchmark.
REST APIs play important roles in enriching the action space of web agents, yet most API-based agents rely on curated and uniform toolsets that do not reflect the complexity of real-world APIs. Building tool-using agents for arbitrary domains remains a major challenge, as it requires reading unstructured API documentation, testing APIs and inferring correct parameters. We propose Doc2Agent, a scalable pipeline to build agents that can call Python-based tools generated from API documentation. Doc2Agent generates executable tools from API documentations and iteratively refines them using a code agent. We evaluate our approach on real-world APIs, WebArena APIs, and research APIs, producing validated tools. We achieved a 55\% relative performance improvement with 90\% lower cost compared to direct API calling on WebArena benchmark. A domain-specific agent built for glycomaterial science further demonstrates the pipeline's adaptability to complex, knowledge-rich tasks. Doc2Agent offers a generalizable solution for building tool agents from unstructured API documentation at scale.