Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 88% confidence
- Finding
- The skill explicitly performs outbound network access to `https://api.signus.ai`, but the metadata shown does not declare corresponding permissions. Undeclared capabilities are dangerous because they weaken operator visibility and policy enforcement, making it easier for a skill to exfiltrate data or access remote services without clear authorization boundaries; the fixed host reduces scope somewhat, but does not remove the governance risk.
