The agents record the answers assistants give to the questions you agree with us, along with the sources shown. They do not need customer lists or personal records to do this. Where an integration needs access, such as read access to Google Search Console, we ask for the narrowest permission that works.
Every captured answer is stored and exportable, which means you can check our work yourself. Every agent action is logged, so you can see what was done, when and on whose approval. Your team can export or delete stored answers on request.
Security is also about process. Your team approves workflows before they go live, and sensitive actions wait for a person to say yes. If your organisation needs a formal review, we can take part in it on the discovery call. Ask us anything that your own policies require.
Retention periods follow your policy, and data is deleted on request. Tell us at the start how long answers should be kept, and we will record it in the blueprint. You can export everything before deletion if you want your own copy. Your data is kept separate from other customers and is not used to train public models. If your organisation requires its own security questionnaire, we are happy to work through it with you.
Yes. Every captured answer is stored and exportable, with the assistant, the date and the sources shown. That lets your team check our reports against the raw text and use the data in your own analysis.