how-to · ai Open this in the console

Read one BE AI investigation.

BE AI runs. A row is one investigation. Opening it does not approve anything.

The list stays put. The detail is the conclusion and the proposals from that run. Approving still happens on the proposal, not here.
01

Open BE AI runs

Open BE AI runs. One finished investigation is listed for Accounts server. The trigger chip reads alert:disk_high. The result is known_growth. The question is about the disk at 92%. Click the row. The filters above are See what BE AI already did.

BE AI runs list with one disk investigation for Accounts server
One finished disk investigation on Accounts server. The row has not been opened.
02

Open the conclusion

The detail names the recipe disk_high, the trigger alert:disk_high, who started it, and that it is done. The classification is known_growth. The summary says the volume is full of SQL backups, and the next step is to approve deleting rotated dumps older than 14 days. Three proposals sit under that: one still waiting, one that failed because the shell is off, and one that already ran. The transcript on this run is empty. Opening the row does not approve a proposal.

Opened BE AI run showing the known_growth conclusion and three proposals
The disk investigation, opened. Conclusion, next step, and three proposals. Nothing was approved.

The list and its filters are See what BE AI already did. A word in the question is Search BE AI runs. One machine is Show BE AI runs for one machine. Approving a proposal is Approve a proposal. What a sub-company user sees, without the token cost, is Read one BE AI investigation on your machines.