comparison-post-writer
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it retrieves and processes untrusted data from external websites using the
web_searchtool without specialized sanitization or boundary markers.\n - Ingestion points: Step 2 of the workflow gathers pricing, features, and user complaints from external sites like G2, Capterra, and Reddit based on user-provided product names.\n
- Boundary markers: The skill does not provide instructions to the agent to treat search results as untrusted or to wrap the retrieved data in protective delimiters to prevent it from being interpreted as instructions.\n
- Capability inventory: The skill has the ability to perform automated web research and generate long-form markdown articles, which provides a pathway for external content to influence the agent's writing.\n
- Sanitization: No explicit instructions are present for the agent to filter, escape, or validate the content retrieved from external search queries before including it in the generated post.
Audit Metadata