Abstract
Performing root cause analysis for cross-domain issues, such as cascading failures in complex, fragmented software ecosystems, can be challenging due to siloed data and technical complexity. To address this, a hierarchical multi-agent orchestration architecture is described. The system can utilize a central orchestrator agent (OA) that receives a diagnostic query, decomposes it into sub-hypotheses, and dynamically routes these tasks to specialized subject matter expert (SME) agents. Each SME agent can be configured with domain-specific knowledge and tools to investigate its assigned task. Through an iterative execution loop, the OA analyzes intermediate findings from the SME agents and can trigger further investigation. This architecture facilitates automated, multi-step deductive reasoning to trace a problem from its symptom to its root cause across disparate technical systems.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Kumar, Sandeep, "Hierarchical Multi-Agent Orchestration for Cross-Domain Root Cause Analysis", Technical Disclosure Commons, (March 25, 2026)
https://www.tdcommons.org/dpubs_series/9616