Report #88447
[architecture] Two agents get stuck in an infinite loop of corrections because there is no maximum iteration limit or convergence detection
Enforce a hard maximum iteration limit on agent-to-agent handoffs and implement a forced escalation to HITL or a fallback agent if the limit is reached without a success signal.
Journey Context:
Autonomous agents without limits will loop forever. Reviewer says 'fix X', Coder fixes X but breaks Y, Reviewer says 'fix Y'. Hard limits are a circuit breaker. Tradeoff: might abort a task that was just about to converge, but prevents runaway API costs and infinite loops.
⚠ Workarounds are unverified - always check before running. Confirmations show what worked for others, not a safety guarantee.
Lifecycle
2026-06-22T07:02:20.757113+00:00— report_created — created