hello-review
Pass
Audited by Gen Agent Trust Hub on May 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local script,
scripts/review-state.mjs, to write review outcomes toartifacts/review.json. This is used for maintaining the state of the review process within the local environment. - [PROMPT_INJECTION]: The skill processes external code content which serves as an ingestion point for potential indirect prompt injection. However, the risk is mitigated by the restricted capabilities of the skill.
- Ingestion points: External code files and PR data provided by the user for review.
- Boundary markers: None explicitly defined in the prompt instructions to isolate untrusted code.
- Capability inventory: Execution of a specific state-writing script (
scripts/review-state.mjs) and file system writes to a designated artifact path. - Sanitization: No specific sanitization or escaping mechanisms are mentioned for the content being reviewed.
- [SAFE]: No malicious patterns, such as hardcoded credentials, obfuscated URLs, or unauthorized remote downloads, were detected. The skill's instructions explicitly direct the agent to check for security vulnerabilities like hardcoded keys and injection flaws.
Audit Metadata