Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 95% confidence
- Finding
- The skill instructs the agent to execute a local Python script, use environment variables, and perform network uploads/downloads, but it declares no permissions or user-visible capability boundaries. That mismatch is dangerous because it can cause the agent to access local files, shell out, and transmit data externally without an explicit trust/consent model, increasing the chance of unintended data exposure.
