Frequently asked questions

Aicentic turns your knowledge base into an AI agent that answers questions, qualifies leads, and takes action -- embedded on your site with a single script tag. It's one agent builder (a knowledge base plus tenant-configured actions) that you point at whatever job you need: sales, support, an internal helpdesk, bookings, and more.

Minutes, not a sprint. Sign up, upload a knowledge base (websites, files, or Q&A pairs), optionally add actions, test it in the built-in simulator, then publish and drop the embed snippet on your site. No engineering work required for the core flow.

No. Training an agent on your content and embedding it is a no-code flow -- a single script tag. Wiring up custom actions (like calling your own booking or order API) is the one part that can involve a developer, since that's connecting to your own systems.

Yes. You can create an account and your first workspace with no credit card required, and try the full build flow -- knowledge base, actions, and the simulator -- before deciding to upgrade.

Free, Starter, Growth, and Pro, plus a custom Enterprise plan for larger deployments. Rates and what's included in each tier are kept current on the pricing page rather than repeated here, since they can change.

Yes, from your account's billing settings. Your plan stays active for the rest of the billing period you've already paid for -- cancelling doesn't cut you off immediately.

Within the first 14 days of a new paid subscription, contact support and we'll review the request -- approved refunds go back to your original payment method. Outside that window, we generally don't provide refunds or credits for partial periods.

You can top up credits directly from billing rather than being cut off mid-conversation -- a top-up is a one-time purchase priced per credit, on top of whatever your plan already includes.

Deleting an agent, a datasource, or your account stops it from being used immediately, but it isn't purged right away -- there's a 30-day grace period so an accidental deletion can still be recovered by contacting support, after which it's permanently deleted.

Routine database backups are encrypted and kept on a rolling 7-day basis for disaster recovery, independent of the 30-day deletion grace period above.

No. Your knowledge base, conversations, and configuration are scoped to your own workspace -- an agent only answers from what you've actually published to it, not from any other tenant's data.

Attach Actions -- external APIs your agent can call mid-conversation -- and it can check an order status, book a meeting, or update a record, not just talk about doing it. Anything that changes something real (a booking, a refund, a cancellation) is built to confirm explicitly with the visitor before it actually runs.

Jira is live today -- tag @Aicentic in an issue comment and the bound agent replies in-thread, with real read/write access to that project's issues. More integrations (Slack, Microsoft Teams, Salesforce, and others) are rolling out -- check the Integrations page in your account for what's currently available.

Only after asking. Every use case involving something billing- or booking-affecting is built to restate exactly what it's about to do and get an explicit yes first -- an agent won't book, cancel, or refund anything on an ambiguous or implied confirmation.

Every connected account's access and refresh tokens are encrypted at rest, never stored in plain text. Each integration also gets its own distinct encryption key, so a problem with one connection type can't expose another's credentials.

Yes -- your workspace's audit log records tenant activity so an administrator can review what changed and by whom, rather than relying on memory or Slack messages.