Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 84% confidence
- Finding
- The skill exposes clear networked capabilities via the declared MCP server and multiple external API endpoints, yet it does not declare any permissions warning consumers that it will communicate off-platform and potentially use environment-backed credentials. In an agent ecosystem, this mismatch is a real security issue because users and orchestrators may grant or execute the skill under the assumption it is low-privilege, while it actually enables outbound requests and authenticated actions such as betting, redemption, and transaction-building.
