Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 93% confidence
- Finding
- The skill clearly instructs the agent to execute shell commands (`bdpan`, `bash install.sh`, `login.sh`, `update.sh`) but does not declare corresponding permissions. This creates a governance and sandboxing gap: the platform or reviewer may treat the skill as lower-risk than it actually is, enabling command execution, package installation, and local state changes without explicit permission scoping.
