OptionalErrorError details (populated when Status is 'Error')
OptionalFinalThe last judge verdict (if the judge was consulted)
Final screenshot captured (base64 PNG)
Final URL the browser was on when the run ended
Terminal status of the run
Complete history of all steps executed
Whether the run achieved its goal
Total wall-clock duration of the run in milliseconds
Total number of steps executed
Complete result of a Computer Use engine run.
Contains the full step history including screenshots (as base64 strings), all actions executed, tool calls, judge verdicts, and timing information.
Layer 1 holds everything in memory. Layer 2 (MJComputerUseEngine) persists screenshots via AIPromptRunMedia entities and logs via MJ's audit system.