This is why it is pretty frustrating. If they actually wanted to audit things, they'd bring in forensic accountants. Instead, they brought in "script kiddies". You aren't finding evidence of crimes after 2 days with SQL access.
Given the time-sensitive nature of the incident, I took the initiative to secure comprehensive system logs, including bash histories, without prior authorization. I was able to identify the root cause: a correctly formatted command executed in an inappropriate context. This proactive approach, while unconventional, was instrumental in mitigating further disruption and expediting resolution.
The root cause was identified, and an effective remediation plan was developed within minutes.
DOGE and their laptops are probably secured by whoever picked them up from Best Buy for Elon.
In my opinion, thinking that AI does a good job of expressing your intent and opinions for you is as flawed as thinking script kiddies with a history of leaking information from their employer deserve a role in breaking and fixing sensitive mission-critical federal systems.