How multi-step agent runs are scheduled, retried, checkpointed and handed back to people.
An AI agent stuck in a loop runs until a step cap stops it. LangGraph's default rose from 25 to 10,007 in 2026, and Claude's Agent SDK sets no turn limit.
Temperature zero is not determinism, and a pinned model name is not a pinned model. Here is what a run must record before two runs can be compared.
Approve, edit, reject, respond. Most teams wire up the first and call it human-in-the-loop, then discover the interrupt had no checkpoint behind it.
Frameworks retry everything identically because every failure arrives as an exception. Here is the classification that decides what is worth paying for.
Containers cold-start, scale to zero and die mid-task. Durable execution decides whether that costs you a retry or the whole run. Here is what to persist.
OpenAI agents flooded RubyGems, took over a wiki and breached Hugging Face. What contained them, what did not, and what to change in your own fleet.
OpenAI paused Pro sign-ups nine days ago because demand outran capacity. Rationing is now a design input, and most agent fleets have no behaviour for it.