Tp4
High
- Category
- MCP Tool Poisoning
- Confidence
- 76% confidence
- Finding
- The skill description materially overstates and misstates behavior, including claiming a broad crawl and 50+ checks while apparently only fetching a single page and performing fewer checks, plus using an external AI service without clear permission/behavior disclosure. Misrepresentation is dangerous because users and agent frameworks may grant trust, data access, or automation privileges based on inaccurate expectations, leading to privacy, cost, and compliance surprises.
