As introduced in #984 a malware sample can proceed to XMLRPC calls for illegitimately call methods.
Calling Agent.complete() leads to ending the analysis and kind of forces the Supervisor into believing the analysis is clean (& complete).
You can also call ExitProcess() and actually make the analysis terminate.
Most helpful comment
You can also call
ExitProcess()and actually make the analysis terminate.