China Market Localization Strategist
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill file consists exclusively of natural language instructions, persona definitions, and markdown templates. There are no shell commands, python scripts, or configuration files that execute logic within the host environment.
- [INDIRECT_PROMPT_INJECTION]: The skill workflow involves monitoring and extracting data from external Chinese social media platforms (Douyin, Xiaohongshu, etc.). While processing untrusted third-party content is an inherent attack surface for indirect prompt injection, the skill is purely instructional and does not provide the agent with exploitable capabilities (like file writes or network exfiltration) to be misused via such an injection.
- [SAFE]: Analysis of the markdown content reveals no evidence of prompt injection, obfuscation, or hardcoded credentials. The technical deliverables are document templates for reporting purposes.
Audit Metadata