Break Node
Exit the enclosing Loop immediately, jumping directly to the loop's "done" port.
LoopBreakSignal flag in the execution memory, which the enclosing Loop node monitors. When this signal is detected, the Loop node stops all further iterations, clears the signal, and fires its done output port. The Break node must always be placed inside a Loop node's body subgraph.
Key Capabilities
- Cleanly exits any active loop without causing an error or leaving the workflow in an inconsistent state.
- Works with the Loop node's standard done port — post-loop nodes are always reached whether the loop ended normally or via Break.
- Preserves the full execution context — all variables set during completed iterations are available after the break.
- Requires zero configuration — the Break node has no properties to set, making it the simplest node in BizFirstAI.
- Typically used after an If Condition node inside the loop body that detects a termination condition.
- Sets
loop_completed_normally = falseon the done port context, allowing post-loop nodes to distinguish between normal completion and early exit.
Business Benefits
Many real-world loop scenarios have a natural early-exit condition. A stock validation loop should stop as soon as it finds an out-of-stock item — there is no need to check the remaining items. A search loop should stop as soon as it finds the first matching record. Without the Break node, workflows would need to process every item even after the termination condition is met, wasting API calls, processing time, and potentially causing unwanted side effects.
The Break node enables a clean "find first" pattern — loop until a condition is satisfied, capture the matching item, and exit. This is used in scenarios such as finding the first applicable pricing rule, locating the first available time slot, or stopping at the first validation failure. All the context built up in the loop body during completed iterations remains available for downstream processing.
Using Break alongside If Condition inside a loop body creates a declarative early-exit mechanism that is visible in the workflow graph. Any developer reviewing the workflow can immediately see that the loop has an early-exit path, what triggers it, and where execution continues after the exit — without needing to understand complex imperative code logic.
Use Cases
Stop Processing Once a Match is Found
A rule engine loops over a list of pricing rules in priority order. As soon as a rule matches the current order's attributes, the matching rule is captured into a variable and the Break node exits the loop. Remaining rules are not evaluated, saving unnecessary processing and ensuring only the highest-priority matching rule applies.
Exit Early on Error Detection
A data import loop processes rows from an uploaded CSV file. If any row fails validation (missing required fields or invalid format), a Break node exits the loop immediately. The done port routes to an error response that returns the offending row number to the user so they can correct and re-upload.
Stop After Finding First Valid Result
A supplier lookup loop queries multiple supplier APIs in order of preference. As soon as one supplier returns an available stock response, the Break node exits the loop. The done port proceeds to place the order with that supplier, without querying the remaining suppliers on the list.
Early Exit on Validation Failure
A multi-step form validation loop iterates over validation rules for a submitted form. The first failing rule triggers the Break, and the loop's done port routes to a validation error response that reports which rule failed. This prevents multiple conflicting error messages from being generated for the same submission.
In This Guide
Configuration
The Break node has no configuration properties. This page covers placement requirements and the LoopBreakSignal mechanism.
Input & Output
The break output port, how the signal reaches the Loop node, and what context is available after the break.
Examples
Practical patterns: find-first, stop-on-error, conditional early exit, and using loop_completed_normally for post-loop routing.