Permissions Overview

CloudPeek only has the access you choose to grant it. Enable only the permissions you are comfortable with: the AI works with whatever you allow, and anything you have not granted is simply unavailable to it.

The permission set is weighted towards read access, which is all the AI needs to triage alerts and investigate. Write permissions are optional, and every state-changing action goes through Human-in-the-Loop approval before it runs. Start read-only if you prefer, and add write access later once you have seen the follow-up actions the AI proposes.

Providers mapped to permissions

CloudPeek has mapped each provider to the permissions it uses, with the use case for every permission, so you can decide exactly what to enable:

ProviderPermissions
Azure / O36557
AWSRead-only, plus opt-in writes
GitHubRead-only, plus optional issue creation
SplunkRead-only search
Elasticsearch / OpenSearchRead-only search
Trend Micro Vision OneRead, plus optional alert write-back
NetskopeRead-only
Threat intelligence servicesRead-only lookups

© 2026 CloudPeek. Agentic AI for high-consequence security operations.