terraform-review
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to read and analyze pull request content, which presents a surface for indirect prompt injection attacks. An attacker could embed malicious instructions within Terraform comments to influence the agent's review. However, the impact is low as the agent is restricted to read-only file system operations.
- [PROMPT_INJECTION]: Ingestion points: Reads PR file content using the Read tool based on Glob patterns.
- [PROMPT_INJECTION]: Boundary markers: Absent. The instructions do not define delimiters to separate untrusted code from the agent's instructions.
- [PROMPT_INJECTION]: Capability inventory: Limited to Read, Glob, and Grep tools, which are read-only and lack network or destructive capabilities.
- [PROMPT_INJECTION]: Sanitization: Absent. The skill does not validate or sanitize code comments before processing.
- [SAFE]: No unauthorized network requests, credential harvesting, or remote code execution patterns were identified.
- [SAFE]: The skill encourages security best practices, such as identifying hardcoded secrets and enforcing least privilege.
Audit Metadata