drift-detector

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires shell execution to run snyk iac describe, terraform, and cloud provider CLI tools such as aws, az, and gcloud. These tools are essential for the skill's stated purpose of infrastructure auditing and drift detection.
  • [DATA_EXFILTRATION]: The drift-remediation.md file contains a shell script example that uses curl to transmit scan results to a user-provided Slack webhook. This is documented as a notification mechanism for administrators and is a standard automation pattern.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources including Terraform state files and cloud environment resource descriptions, which constitutes a potential surface for indirect prompt injection. * Ingestion points: Terraform state files and cloud provider API responses. * Boundary markers: The instructions do not provide explicit delimiters to separate untrusted infrastructure data from agent instructions. * Capability inventory: The skill utilizes shell execution, file reading, and file writing as defined in the frontmatter. * Sanitization: No specific sanitization or validation logic is defined for the ingested infrastructure data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 02:46 PM