How to stop AI Agent and Agentic Workflows in in-progress state that were unintentionally triggeredIssue A test Agentic Workflow in the Dev environment was unintentionally triggered by another project, which resulted in the workflow launching hundreds of jobs simultaneously. This unexpected activation caused a significant consumption of system assists, impacting overall system performance. Although the triggers responsible for initiating these jobs were identified and disabled to prevent further launches, a substantial number of workflow jobs remain active, either continuing to run or stuck in an in-progress state without completion. To mitigate any additional resource drain and avoid unnecessary processing, the current request is to immediately stop and cancel all remaining in-progress jobs associated with this workflow.Releasen/aResolution1. Load the URL: https://INSTANCE_NAME.service-now.com/cancel_my_transaction.do 2. Execution Plans: (Only 1 Abandoned record) Access the execution plans by navigating to the following URL: https://INSTANCE_NAME.service-now.com/sn_aia_execution_plan_list.do?sysparm_query=GROUPBYstate. This will allow you to view the current execution plans and identify any that are abandoned or require attention.3. Clear Stuck Subflow Contexts - Navigate to: All → Flow Designer → Action Executions - Apply a filter for records where the State is set to Pending. This helps isolate subflow executions that have not progressed as expected.- Review these records carefully and cancel or delete them if it is safe to do so, ensuring that no critical processes are interrupted.4. Cancel Orphaned Contexts: Sometimes executions remain stuck even after triggers are disabled, which can cause workflow delays or errors.- Navigate to: All → System Scheduler → Contexts - Use filters to locate contexts associated with the specific Agentic workflow name.- Focus on records with states marked as "In Progress" or "Waiting for input" to identify those that are potentially orphaned.- Cancel or delete these records if it is safe, to clear the backlog and restore normal workflow operation.