Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 95% confidence
- Finding
- The skill declares sensitive capabilities in metadata (`env` secrets and external network use) and documents automatic calls to third-party APIs, but it does not expose an explicit permission model for users approving those actions. This creates a transparency and consent gap: the skill can access credentials and send camera-derived data off host without a formal permissions declaration beyond descriptive text.
