Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 88% confidence
- Finding
- The skill explicitly instructs the user to run a Python script that fetches live public quotes from multiple exchanges, which requires outbound network access, yet no corresponding permission is declared. This creates a transparency and policy gap: the skill's real capabilities are broader than its declared security model, making review, sandboxing, and user trust weaker.
