Configuration for timing out node attempts.
A timeout applies to a single attempt of a node/task. When a node has a RetryPolicy, the timer resets for each retry attempt.
Timeouts are expressed in milliseconds, matching other LangGraph.js
durations (e.g. RetryPolicy intervals and stepTimeout).
Cooperative cancellation: timeouts rely on aborting the node's
AbortSignal and dropping its buffered writes. A node that ignores its
signal and performs blocking work cannot be interrupted mid-operation, but
its writes are still discarded and NodeTimeoutError is raised.
Maximum time (in milliseconds) a single node attempt may go without observable progress before timing out.
Refreshed by writes, custom stream writer calls, child-task scheduling,
LangChain callback events emitted under the node's run, and explicit
runtime.heartbeat() calls (see TimeoutPolicy.refreshOn).
Which signals refresh TimeoutPolicy.idleTimeout.
"auto" (default): refreshes on standard graph progress signals and
explicit heartbeats."heartbeat": refreshes only on explicit runtime.heartbeat() calls.Hard wall-clock cap (in milliseconds) for a single node attempt.
This timeout is never refreshed by progress signals or
runtime.heartbeat().