cloud-security

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and use several industry-standard security tools from reputable organizations. This includes pacu and CloudGoat from Rhino Security Labs, ScoutSuite from NCC Group, trivy from Aqua Security, and kubeaudit from Shopify. These tools are well-known in the cybersecurity industry and their usage is consistent with the skill's primary purpose of cloud security assessment.
  • [COMMAND_EXECUTION]: The skill makes extensive use of the Bash tool to execute cloud service provider CLIs (aws, az, gcloud) and Kubernetes management tools (kubectl). These operations are necessary for the skill's function and are directed toward the user's cloud environment for auditing purposes.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes data from external cloud environments, which acts as a potential attack surface.
  • Ingestion points: Data is ingested via Cloud API responses (IAM policy documents, bucket listings, resource tags, and CloudTrail logs) processed by scripts in guides/cloud_security_complete_guide.md and payloads.md.
  • Boundary markers: The provided analysis scripts do not implement explicit boundary markers or "ignore instructions" tags for the processed cloud data.
  • Capability inventory: The skill has access to Bash, Write, Edit, and WebFetch tools, providing a broad range of post-ingestion capabilities.
  • Sanitization: There is no evidence of specific sanitization or validation of cloud resource identifiers or metadata before they are processed by the internal Python auditing scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 04:36 PM