skills/dtprj/dongting/idea-mcp/Gen Agent Trust Hub

idea-mcp

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The documentation describes how to interact with the IntelliJ IDEA MCP plugin for Java development tasks such as code navigation, searching, and refactoring. It does not contain any executable scripts or hidden commands.
  • [COMMAND_EXECUTION]: The documentation references a tool named 'execute_terminal_command'. It provides informative guidance on its behavior (not using a shell by default) and recommends using standard bash tools instead for shell-specific features, which is a neutral and helpful instruction.
  • [INDIRECT_PROMPT_INJECTION]: The skill mentions tools that read project files, which is an inherent attack surface for indirect prompt injection. However, as a documentation-only file, it does not implement any unsafe interpolation or sanitization bypasses itself.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: There are no credentials, sensitive file paths, or network exfiltration patterns present. The example paths provided are generic and for illustrative purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:10 PM