1688-marketing-skill
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions provide a method to clone a source repository from GitHub (
next-1688/1688-marketing) and install dependencies usingpip. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data from the 1688 platform, including merchant activities, pricing data, and business opportunity recommendations. This creates an attack surface where malicious instructions embedded in external platform data could potentially influence agent behavior. The skill documents the use of an
EnrollmentValidatorand data validation best practices as mitigation measures.
Audit Metadata