Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 96% confidence
- Finding
- The skill declares no permissions, but its documented behavior clearly includes reading and writing local files, accessing environment-based secrets, and making network requests. This mismatch is dangerous because it prevents informed consent and review; a user or platform may invoke the skill expecting a simple lookup tool when it can also access credentials, persist data locally, and contact remote services.
