Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 88% confidence
- Finding
- The skill explicitly requires environment variables and network access in metadata, but no declared permissions model is presented to constrain or disclose those capabilities. In an agent environment, undeclared sensitive capabilities increase the chance of secret exposure or unexpected outbound requests because operators may not realize the skill can read an API key and transmit data externally.
