Terraform Plan Reviewer
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown documentation and YAML metadata. It does not include any executable scripts (.py, .js, .sh), binaries, or tool implementations.
- [INDIRECT_PROMPT_INJECTION]: The skill describes a workflow for parsing external data (Terraform plan JSON output). This constitutes an attack surface for indirect prompt injection, as an attacker controlling the infrastructure code could potentially embed instructions within the plan output to influence the agent. However, as no implementation code is provided, this is a theoretical risk inherent to the task rather than a vulnerability in the skill itself.
- Ingestion points: Terraform plan JSON output (referenced in SKILL.md).
- Boundary markers: None specified in the documentation.
- Capability inventory: Identifying destructive changes and cost implications.
- Sanitization: None described in the provided instructions.
- [EXTERNAL_DOWNLOADS]: The documentation references resources from the official HashiCorp GitHub repository and developer website. These are trusted sources for infrastructure tooling documentation.
Audit Metadata