If incidents live as WordPress posts (a status CPT or block-based posts) SleekAI reads them natively, including the status field, started_at, resolved_at, and component scope. For externally hosted status pages (Statuspage, Better Stack) you have two options: mirror incidents into a WordPress CPT, or feed the JSON status endpoint into the system prompt at startup so the bot can quote the current state.

Yes. Map a 'status' field on each incident post (active, investigating, monitoring, resolved) and the system prompt instructs the model to lead with active incidents first, falling back to 'all systems operational' when none exist. The bot can also bound 'recent' to a configurable window, the last seven days by default, to keep replies sensibly scoped.

As current as your status feed. SleekAI reads posts on every request, so a new update published 30 seconds ago is in the next reply. There is no scheduled re-index step or fine-tune retraining, the bot stays in sync with your live feed automatically, which matters most during an incident.

Yes, if your incident posts carry a component taxonomy (API, webhooks, dashboard, ingestion). The bot can answer scoped questions like 'is webhooks delivery degraded' or 'is the dashboard slow' by filtering on the taxonomy term, instead of dumping the full feed. For very large estates, multibot can run a per-product status bot.

Link each resolved incident to its postmortem URL via an ACF field, and the bot will offer the postmortem on the next turn when a customer asks 'what happened'. That turns a worried-customer question into a transparent retrospective, often the highest-leverage trust gesture an SRE team can make.

Indirectly, via two practical patterns. Mirror incidents from the third-party status tool into a WordPress CPT on every state change (webhook in, post out), so your WordPress install is the canonical source for the bot. Or, feed the JSON status endpoint into the system prompt at startup, so the bot reads live state without persisting it. Both work, the choice depends on whether you want a queryable archive.

SleekAI is a one-time WordPress plugin license, not a per-conversation SaaS subscription. You bring your own OpenAI, Anthropic, Google, or OpenRouter API key, so you pay the provider directly at standard API rates. There is no token markup, no resolved-conversation fee, and no monthly seat cap, which keeps cost predictable even during a high-traffic incident.

The system prompt explicitly forbids invented incidents and tells the model to default to 'all systems operational' when no active incident posts are in context. Retrieval keeps it grounded, if no recent incidents match, the model defers cleanly. Conversation logs let you spot any drift, the rare case is almost always a malformed post that retrieval did not surface, fix the post and the answer is correct.