terraform-infra-architect

Warn

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local shell script (scripts/scaffold-repo.sh) to automate project creation. This script performs dynamic file generation, including Makefiles and GitHub Actions workflows, by interpolating user-controlled variables. Without rigorous validation, these inputs could influence the resulting executable automation logic.
  • [COMMAND_EXECUTION]: The generated project contains a Makefile and CI/CD pipelines that execute shell commands based on parameters defined during scaffolding. This creates an indirect path where untrusted user input influences command-line operations.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface due to its broad capabilities. 1. Ingestion points: User responses to configuration questions in SKILL.md. 2. Boundary markers: Absent; user input is directly interpolated into file templates. 3. Capability inventory: File writing, directory creation, and execution of shell scripts. 4. Sanitization: Absent; the script performs no escaping or validation of user-provided strings before processing.
  • [EXTERNAL_DOWNLOADS]: The generated GitHub Actions workflows reference official setup actions from HashiCorp and AWS. These downloads are performed from trusted repositories and official service providers.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 13, 2026, 02:00 AM
Security Audit — agent-trust-hub — terraform-infra-architect