Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 92% confidence
- Finding
- The skill clearly instructs users to run a Python script against arbitrary URLs, which implies outbound network access, yet no corresponding permission is declared in the skill metadata. Undeclared network capability is dangerous because it hides the real trust boundary from users and platform policy enforcement, making SSRF-style access, unexpected external requests, or data exfiltration harder to detect and govern.
