Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 92% confidence
- Finding
- The skill advertises environment-variable and network-driven behavior but does not declare corresponding permissions, which prevents hosts or reviewers from understanding that it can access secrets and communicate with external services. In an agent ecosystem, undeclared capabilities materially increase the risk of secret exposure or unintended outbound actions because operators may install the skill under incomplete assumptions.
