A production agentic system sometimes enters infinite loops, calling the same tool repeatedly with identical parameters. QA testing confirms this happens in roughly 3% of runs. What architectural safeguard should be implemented?

Infinite loop detection and prevention must be implemented in the orchestration layer (code), not trusted to Claude's instruction-following alone. The safeguard has two components: (1) active loop detection — compare each tool call against recent call history and detect identical (tool, parameters) pairs; (2) maximum iteration count — enforce a hard cap (e.g., 50 steps) with human escalation or graceful failure when reached. When a loop is detected, inject a guidance message: 'You have called [tool] with these parameters 3 times without progress. Consider an alternative approach.' Option A (larger context) is irrelevant — Claude can see its call history in context and still loops; the issue is a reasoning failure, not context blindness. Option C (temperature 0) makes loops more likely, not less — a deterministic model will repeat the same mistake every time with zero variation. Option B (instruction alone) is insufficient; agentic loops occur because the model genuinely believes each next call will differ.