dropship-product-research-sop
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process external, untrusted content such as Amazon reviews and Facebook advertisement copy. This creates an ingestion surface where instructions embedded in the external text could potentially influence agent behavior. 1. Ingestion points: Step 3 (Amazon reviews) and Step 6 (Facebook ads) in SKILL.md. 2. Boundary markers: The provided prompt templates lack explicit delimiters or 'ignore instructions' warnings for the ingested content. 3. Capability inventory: The skill does not use any scripts, tools, or shell commands. 4. Sanitization: No sanitization or validation of the pasted content is recommended in the SOP.
- [NO_CODE]: The skill consists exclusively of markdown documentation and instructions, without any accompanying scripts, executables, or tool configurations.
Audit Metadata