DYNAMIC LLM AGENT METADATA MANIFEST-BASED DISCOVERY OF AGENTS IN AN LLM AGENTIC APPLICATION PLATFORM
Abstract
Within Large Language Models (LLM) environments, agents can be provided to perform a variety of complex tasks. However, with many agents being developed, it can sometimes be difficult to discover or select an appropriate agent to perform a desired task. Techniques presented herein can be used to aid a developer in identifying the best artificial intelligence (AI) agent to utilize for a given set of criteria. Based on the proposed techniques, a low code / no code agentic infra platform can be provided that may facilitate the efficient discovery of agents in an agnostic manner regardless of the underlying framework in which agents are provided.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Sheriff, Akram, "DYNAMIC LLM AGENT METADATA MANIFEST-BASED DISCOVERY OF AGENTS IN AN LLM AGENTIC APPLICATION PLATFORM", Technical Disclosure Commons, (November 12, 2024)
https://www.tdcommons.org/dpubs_series/7522