Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 89% confidence
- Finding
- The skill instructs the agent to execute local shell commands and read local files/scripts, but there is no declared permission model or limitation around those capabilities. In an agent environment, undeclared shell and file access increases the chance of unintended local inspection or command execution, especially because the skill treats those actions as mandatory preflight steps.
