review-reply-method
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data in the form of customer reviews provided by the user.
- Ingestion points: The review text is provided as input via
SKILL.md(Process Step 1). - Boundary markers: The instructions do not define clear delimiters or use specialized prompting techniques (such as XML tags or "ignore embedded instructions" headers) to separate the untrusted review content from the skill's instructions.
- Capability inventory: The skill has no dangerous capabilities; it lacks tool access for network operations, file system writes, or shell command execution. Its scope is limited to text analysis and response generation.
- Sanitization: There is no evidence of input validation or sanitization for the review text before it is processed by the agent.
Audit Metadata