Skip to main content
Yes. She keeps memory of durable facts between conversations, and any open backlog item stays with her until it’s done, cancelled, or you tell her otherwise. See Memory and Abbie’s backlog.
No. Everyone gets their own Abbie: your conversations, your personal backlog and your personal rapport are yours alone. Team-scoped backlog items are the exception, since those are deliberately shared. See Personal vs team backlog.
Nothing is lost. A backlog item she’s working on keeps going, and you can pick the conversation back up later from your session list or from Slack, depending on where it started.
The web chat works on its own; Slack is optional, and only worth connecting if you want Abbie reachable where your team already talks. See Slack.
No. Turning off check-ins, follow-ups and suggestions only stops Abbie from starting a conversation herself; she answers exactly the same whenever you talk to her first.
Yes, from the Synapse Memory tab in her settings. Once deleted, she won’t quietly re-learn and re-add the same fact later. See Memory.
They show the same backlog. The Backlog page is the full management view: filtering, scheduling, promoting items, seeing your team’s backlog. The in-chat panel is a lighter view for quick actions without leaving the conversation. See The Backlog page and The in-chat panel.
No. Assistants are bots you build and configure yourself, each for a particular job. Abbie is the single coworker already built into QuivaWorks — everyone on the account gets their own, with their own memory, backlog and rapport.
Under Monitoring → Usage & Spend, which covers consumption across the whole account rather than Abbie alone.
Yes. Teams needs more setup than Slack — admin consent alone does not put her in Teams, the app package also has to reach your Teams app catalog. See Microsoft Teams.
Abbie can only act through her own dedicated login or your personal connection, never a shared, team-level connection configured for other purposes. If neither of those exists yet for a given integration, she’ll ask. See Credential identity explained.
If, after her own internal review, an answer still scores poorly, Abbie says so up front rather than presenting a shaky answer as certain. See Confidence and verification.