Yes. MalCare stores scan results, hardening flags, and login protection events in WordPress options and its own tables, which SleekAI can be configured to read as data sources. The bot's system message exposes those values as named variables, so the model has structured data to answer with rather than generic security copy. The wizard handles the mapping in a few minutes.

Not unless you configure it that way. Display conditions on user role, capability, logged-in state, and URL pattern keep the security bot scoped to administrators inside wp-admin. Any public chatbot you run on the same site is a separate bot under multibot, with its own data sources and no access to MalCare tables.

The bot can read whatever MalCare writes to your database, so anything visible to you in the WP-side MalCare dashboard is fair game. Some advanced fields, like deep scan details, only exist on paid plans. The basic flow of scan timestamps, blocked logins, and hardening flags works regardless of tier.

By default, SleekAI is read-only against MalCare data. It surfaces state and explains what is pending. If you want write actions, you can wire a custom PHP callback into a SleekAI tool and have the bot call it under a tight permission check, but most teams treat the chatbot as a read layer and keep destructive actions inside the MalCare UI.

Yes. Conversation logging is on by default per chatbot, with user messages, assistant replies, model name, token usage, and the page URL stored in WordPress. For a security bot, that audit trail matters: the lead can see which staff asked what, when, and which model produced the answer.

Yes, as long as MalCare's login protection log is enabled and SleekAI is wired to read it. Ask for blocked attempts in the last 24 hours, the top source IPs or ranges, and the times of day under attack, and the bot rolls the rows up into a short summary instead of a screen of raw entries.

A fast model is fine for routine reads like scan status and blocked-login counts. For longer triage prompts that summarise an incident across multiple data sources, a stronger model from OpenAI, Anthropic, Google, or OpenRouter gives more useful synthesis. SleekAI lets you pick the model per chatbot, so you can keep cost in line with the workload.

Each SleekAI install runs its own chatbots with its own data sources, so the configuration is per site. You can export and import the chatbot config across sites and re-point the data sources, which is useful when an agency runs the same security bot across several client sites with the same MalCare setup.