qinghu-bilibili-social
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Bilibili, specifically through the analysis of video comments and search results.
- Ingestion points: Data is retrieved from external sources using the
bilibili_video_commentsandbilibili_search_videotools defined in the skill. - Boundary markers: The instructions do not define specific delimiters or "ignore instructions" markers to safeguard the model against malicious instructions potentially embedded in social media comments.
- Capability inventory: The skill possesses network communication capabilities to interact with the vendor API at
https://www.iqinghu.com/api/desktop/qh/mcp. - Sanitization: The provided instructions do not mention any specific sanitization or filtering logic for the external content before it is processed by the agent.
Audit Metadata