Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 88% confidence
- Finding
- The skill invokes a Python script that uses environment variables, reads a local secrets file, and performs network access, but the skill metadata does not declare those capabilities. This is dangerous because it hides the actual trust boundary and can surprise policy enforcement or reviewers, especially since the skill can access credentials and make outbound requests.
