Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 92% confidence
- Finding
- The skill instructs the agent to execute a shell command (`python3 scripts/search.py "<关键词>"`) but does not declare any permissions for shell/code execution. That creates a capability mismatch: a host may either block execution unpredictably or, worse, permit undeclared command execution, increasing the attack surface. In this context the command template includes user-influenced input, so if the implementation later interpolates it unsafely, it could become command injection or unauthorized local file access.
