From Agent Traces to Trust: Evidence Tracing and Execution Provenance in LLM Agents

ArXi:2606.04990v1 Announce Type: cross Large language model (LLM)-based agents increasingly solve complex tasks by interacting with external tools, retrieval systems, memory modules, environments, and other agents. These capabilities expand agent autonomy, but also make agent behavior harder to verify, debug, and audit. Final-answer accuracy alone cannot explain how an output was produced, which evidence ed each claim, whether tool calls were justified, how memory influenced later decisions, or where execution failures originated.