Workflow Execution

You can automate multi-step tasks by defining your processes in YAML. The execution engine resolves the steps into a Directed Acyclic Graph (DAG) to orchestrate sequential and parallel execution while managing the dynamic context.

Context management passes data between steps using a dynamic context object. This allows the execution engine to substitute variables and maintain state across both sequential and parallel tasks within the resolved DAG.

Yes, you can resume workflow execution directly from a failed step. This error handling capability prevents manual intervention and ensures reliable task completion without re-running the entire multi-step process.

Yes, workflow execution logs results to Notion databases. It records the execution status, dynamic context, and step results automatically, ensuring you can track the progress of complex automated tasks.

The best way to handle errors in task execution is using an engine that implements robust error detection and recovery strategies. This approach manages data flow gracefully and supports workflow resumption from failed steps.

You can define step conditions when orchestrating a Directed Acyclic Graph (DAG). These conditions, combined with context substitution, allow the execution engine to evaluate whether specific tasks should run during the workflow.