terraform-review
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs static analysis of Terraform files to detect security risks such as hardcoded secrets, overly permissive security groups, and lack of encryption. It is restricted to a small set of safe tools (Read, Glob, Grep) and lacks network or write permissions.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted data as it is designed to process arbitrary Terraform configuration files.
- Ingestion points: Local .tf files accessed via the Read and Grep tools as part of the PR review process.
- Boundary markers: Absent; the skill instructions do not specify delimiters for separating external file content from the agent's internal reasoning.
- Capability inventory: Limited to file system discovery and reading. No network access (curl, fetch), dynamic execution (eval, exec), or file system write capabilities are granted in the allowed_tools manifest.
- Sanitization: The skill processes file content without explicit escaping or sanitization logic, relying on model-level constraints.
Audit Metadata