agency-china-market-localization-strategist
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves processing untrusted external data, which creates a potential surface for indirect prompt injection.
- Ingestion points: The strategist is instructed to aggregate data from at least 7 platforms (Douyin, Bilibili, Weibo, etc.) in SKILL.md and perform "Comment Track" analysis on user-generated content.
- Boundary markers: The instructions do not define clear delimiters or specialized prompts to help the agent distinguish between its instructions and potentially malicious content embedded in the ingested trend data or comments.
- Capability inventory: The agent is designed to generate executable strategies, checklists, and content templates based on the data it processes as described in SKILL.md.
- Sanitization: There is no evidence of instructions to sanitize or filter the external data for prompt injection patterns before analysis.
Audit Metadata