offensive-cloud-role
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions to execute cloud management commands such as
aws sts get-caller-identityto establish principal identity within a cloud environment. - [PRIVILEGE_ESCALATION]: The instructions encourage the agent to identify paths for lateral movement and role assumption, specifically questioning if a workload identity allows movement into Kubernetes clusters or if a role can assume another role.
- [DATA_EXFILTRATION]: The skill directs the agent to search for sensitive information and credentials stored in cloud services like AWS S3 buckets, Azure Blobs, Secrets Manager, and Parameter Store.
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill instructs the agent to read and enumerate data from external cloud storage resources (S3 buckets, Azure Blobs).
- Boundary markers: Absent. No instructions are provided to the agent to treat content from these external resources as untrusted or to ignore instructions contained within them.
- Capability inventory: The agent is directed to use cloud CLI tools which possess capabilities for file writing, network communication, and resource management.
- Sanitization: Absent. The skill does not define methods to filter or escape content retrieved from external cloud storage.
Audit Metadata