Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 88% confidence
- Finding
- The skill instructs the agent to invoke a shell script (`scripts/linear.sh`) and handle a credential file, but it does not declare any permissions despite clearly requiring shell execution. This creates a trust and containment gap: an orchestration system may expose shell capability implicitly, allowing command execution and access to local secrets without an explicit permission boundary or user awareness.
