skills/transilienceai/shasta/scan/Gen Agent Trust Hub

scan

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes Python command-line execution to perform complex operations including database management, cloud credential validation, and multi-cloud compliance scanning using the internal 'shasta' library.
  • [SAFE]: All identified operations, including file access and cloud API interactions, are consistent with the skill's stated purpose of providing security compliance audits. The skill relies on vendor-specific internal modules for its logic.
  • [PROMPT_INJECTION]: The skill processes findings from external cloud environments which presents a surface for indirect prompt injection if resource names or metadata contain malicious instructions. This is a common attack surface for data-processing tools.
  • [Ingestion points]: External data is ingested from cloud provider APIs via 'run_full_scan' and local configuration from 'shasta.config.json'.
  • [Boundary markers]: No explicit delimiters are used to separate scanned findings from agent instructions.
  • [Capability inventory]: The skill can execute arbitrary Python commands and write report files to the local filesystem.
  • [Sanitization]: No explicit sanitization of scan results is described in the instruction set.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 12:15 PM
Security Audit — agent-trust-hub — scan