Question 1

Can Ana query PagerDuty analytics?

Accepted Answer

Yes. PagerDuty's analytics API exposes incident trends, MTTR, MTTA, and responder performance. Ana pulls these metrics and joins them with warehouse data to answer questions like "how did our MTTR correlate with customer churn last quarter?" in a single analysis.

Question 2

Does Ana read PagerDuty schedules?

Accepted Answer

Yes. The connector lets Ana read on-call schedules, override history, and coverage gaps. This is useful for analyses that tie incident load to on-call rotations or that surface coverage gaps correlated with warehouse-reported customer issues.

Question 3

How are PagerDuty keys secured?

Accepted Answer

API keys are encrypted in TextQL's credential vault and injected by our egress proxy at the network layer. Ana's Python sandbox never sees the raw key value. Keys can be rotated any time from the TextQL admin console without changing any analysis prompts.

Question 4

Can I use a read-only key?

Accepted Answer

Yes, and we recommend it. PagerDuty offers read-only API keys that cannot modify incidents or schedules. Create a dedicated read-only key for TextQL to enforce a clean boundary and avoid any risk of Ana accidentally changing operational data.

Question 5

Can Ana create PagerDuty incidents?

Accepted Answer

Yes with write permissions. By default the connector is read-only. Admins can grant write permissions to let Ana create incidents, add notes, or trigger escalations from analysis results. Writes are audit-logged and can be restricted to specific services.