detecting-infrastructure-drift

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses Bash tools scoped to terraform, aws, and gcloud commands to interact with infrastructure. It also references a local script, scripts/drift-check.sh, which automates plan and preview actions for IaC tools like Terraform and Pulumi.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process IaC configuration data from external files.
  • Ingestion points: Reads IaC configuration from {baseDir}/terraform or equivalent directories (SKILL.md).
  • Boundary markers: No specific instructions or delimiters are used to ensure the agent ignores potentially malicious instructions embedded within the processed configuration files.
  • Capability inventory: The skill is granted Write, Edit, and Bash execution capabilities (SKILL.md).
  • Sanitization: There is no documented mechanism for validating or sanitizing the content of the IaC files before the agent processes them.
  • [SAFE]: The skill's behavior is entirely aligned with its declared functionality. External links point to official documentation for well-known services (Terraform and AWS), and all operations occur within the expected scope of infrastructure management. No signs of exfiltration or malicious intent were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 11:50 AM