Tp4
- Category
- MCP Tool Poisoning
- Confidence
- 97% confidence
- Finding
The skill appears to perform external HTTP requests, manage auth headers/tokens, resolve user identities, and read/write local workspace data, none of which are clearly reflected in the top-level purpose. This is especially dangerous because the inputs are health-related videos and historical reports, making undisclosed data transfer and identity linkage higher-risk than in a generic media tool.
- Content
