product-detail-page-conversion-review
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from product descriptions, competitor info, and user files.
- Ingestion points: Data enters via the parameters productPageText, competitorLinks, and customerConcerns, as well as via the pageFile upload field defined in references/form-schema.json.
- Boundary markers: The instructions do not employ delimiters or specific warnings to ignore instructions embedded within the processed content.
- Capability inventory: The skill includes a local runner script (scripts/run.py) capable of performing network requests.
- Sanitization: There is no evidence of data sanitization or filtering applied to external inputs before processing.
- [COMMAND_EXECUTION]: The skill includes a Python script (scripts/run.py) designed to be executed by the user to send request data to the vendor's API infrastructure.
Audit Metadata