cloud-iam-deep

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for processing untrusted cloud credentials and tokens which could serve as a vector for malicious instructions designed to influence the agent's behavior during interaction with cloud APIs.
  • Ingestion points: The skill processes untrusted credentials (AWS keys, GCP service account JSON, and K8s tokens) discovered from external sources like code repositories, APKs, or breach corpora (SKILL.md).
  • Boundary markers: There are no explicit boundary markers or instructions provided to the agent to treat the contents of these credentials as data rather than potential instructions.
  • Capability inventory: The skill utilizes powerful command-line interfaces (aws, az, gcloud, kubectl) and curl to perform network operations and manipulate cloud resources across all provided modules.
  • Sanitization: The instructions do not include validation or sanitization steps for the input credentials before they are interpolated into shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 05:32 AM
Security Audit — agent-trust-hub — cloud-iam-deep