Starts a new incident episode and returns the observation. No grading happens yet.
Manual Evaluation
Interactive playground
This page is a browser version of the OpenEnv flow. Reset starts one evaluation episode, Step submits one agent answer, and the result shows the reward returned by the grader.
Check the incident, expected field, allowed values, and context.
Send one answer. The backend grades it and prints a terminal log.
Step 1
Start a session
Session
Not started
Pick a preset or enter a task and ticket manually.
Step 2
Submit an action
The playground automatically maps your choice to `severity`, `root_cause`, or `action`. If you choose a known ticket, it also sets the matching task type for you.
Observation Brief
What the agent sees
Start a session to load the incident alert.
Allowed values
Start a session first
Context signals
No context loaded yet.
Grader Result
What the reward means
Waiting for step
Reason
Submit a step to see the deterministic grader explanation.
Observation
Latest reset payload
No observation yet.
Result
Latest step payload
No step submitted yet.