Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 93% confidence
- Finding
- The skill declares required credentials and uses code that accesses environment variables and makes outbound network requests, but the metadata only specifies allowed tools and does not clearly declare these sensitive capabilities as permissions. This creates a transparency and governance gap: an agent or reviewer may underestimate that the skill can access secrets and transmit authenticated requests to an external service.
