Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 94% confidence
- Finding
- The skill instructs the agent to execute local Python tools that fetch remote news and market data, implying shell execution, network access, and likely file reads, yet no permissions are declared. This creates a transparency and policy-enforcement gap: the skill can perform sensitive actions users or the platform may not expect, making misuse or overreach harder to control.
