The Lab Options in the Console Menu allow you to visualize how Agents are executed, what data they generate, and which Reasoning Strategies are available for reuse or adjustment. This section also provides access to all artifacts produced during executions—such as text outputs, Markdown files, or translations—so you can review, download, and reuse them without rerunning the original Agentic Process.
Within The Lab, you can access the following:
- Jobs: Displays the Agent name, the executed task, the subject or input, cost, duration, and final status. You can use this section to confirm that an Agent ran successfully, identify errors, and analyze execution performance and costs.
- Processes: Shows the instances of all the Agentic Processes that have been executed. Each record indicates the version, subject, state (running, completed, or canceled), cost, and timestamps. Use this section to monitor and analyze multi-step flows, verify that all stages of a process executed as intended, or find where an orchestration was interrupted.
- Files: Lists all artifacts created or uploaded as a result of Jobs or Processes. You can filter by name or type, identify the source that generated each file, check creation dates, and download outputs. Use this section to access results such as Markdown documents, translations, or text files generated by Agents, without needing to re-execute them.
- Reasoning Strategy: Allows you to create, upload, and manage the Reasoning Strategies that guide how your Agents process information. You can import strategies from external sources or create new ones within the Console. Each Reasoning Strategy includes details like name, type, access scope, and current status.
At the top right of the Console Header, use the Project Selector to choose the specific Project you want to view. Once selected, all information displayed in The Lab—including Jobs, Processes, Files, and Reasoning Strategies—is automatically filtered to show only the executions and data related to that Project.
Since version 2025-04.