Decide
Branch using deterministic operators or an explicit AI rule.
Workflow creators.
Overview#
Branch using deterministic operators or an explicit AI rule.
Use this block when its responsibility is clear in the graph. Give it a name that describes the work it performs, then test it with representative input before publishing.
Working examples#
These are complete starting points, not decorative diagrams. Every graph is also available in the main template library and is checked by the workflow compiler.
1. Route urgent support cases#
You will build: Urgent cases separated from normal review.
Uses an explicit branch instead of burying routing in a prompt. A deterministic equality check is stable across runs.
Open this template and adapt the labels, prompts, or credentials to your own workflow.
2. Require approval for high spend#
You will build: High-spend requests routed for approval.
Makes the control visible in the graph. A numeric greater-than rule is auditable and model-free.
Open this template and adapt the labels, prompts, or credentials to your own workflow.
3. Check publication readiness#
You will build: Ready work separated from incomplete work.
Stops incomplete content before publication steps. The false branch remains visible instead of silently dropping work.
Open this template and adapt the labels, prompts, or credentials to your own workflow.
Ports#
- Input: Value (
input):any, required, single. - Output: True Branch (
true):any, optional, single. - Output: False Branch (
false):any, optional, single.
Inspector#
- AI Rule (optional): key
humanCondition; textarea; default:"". - Condition Type: key
operator; select; options:truthy,falsy,equals,notEquals,gt,lt; default:"truthy". - Compare Value: key
compareValue; text; default:"". - condition editor: manages
operator,compareValue,humanConditionwith edit, validate, preview support.
Behavior#
- Input requirement: Required input ports:
input. - Execution: Handled by the handler using
condition. - Timeout: No node-level timeout is defined.
Cost#
Uses a language-model provider and contributes model usage to the run cost.
Security#
- Uses a runner-provided model API key. Secret values are not part of the published graph.
- The node reads from an external service without intentionally mutating it.
Retry#
The runtime may retry this node according to the compiled execution policy.
Production checklist#
- Test the happy path and one malformed or empty input.
- Keep credentials out of prompts and published graph configuration.
- Set a timeout and retry policy that matches the operation's side effects.
- Name the final result so a runner can tell what success looks like.
Was this useful?
Your response helps us improve the documentation.