Agents are tools designed to perform a specific task in the context of a greater workflow. They may or may not call AI models themselves, but most importantly, they accomplish a goal when given a well-defined task that is within their scope and return a well-formatted output. This allows larger AI-driven workflows to plan and execute workflows using these tools.
URSA is an agentic workflow toolkit for science, created by LANL, that contains a bunch of examples of what agents do. Here’s an arxiv agent: https://github.com/lanl/ursa/blob/main/docs/arxiv_agent.md.
- it relies on Chroma as its vector database.