Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 90% confidence
- Finding
- The skill explicitly describes reading an authentication token from environment variables and sending user query parameters to an external proxy/API, which indicates effective use of env and network capabilities despite no declared permissions. This is dangerous because undeclared capabilities reduce transparency and reviewability, making it harder for operators and users to understand what data leaves the system and what secrets the skill can access.
