cross-platform-compatibility
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides numerous templates and reference implementations across multiple files (references/shell-commands.md, references/process-management.md) for executing shell commands and managing system processes using Node.js child_process.exec and child_process.spawn. These are standard developer tools, but they represent a powerful capability.
- [INDIRECT_PROMPT_INJECTION]: The reference code in references/shell-commands.md and references/process-management.md contains significant vulnerability surfaces for command injection. Specifically, functions like openFile, listFiles, and findProcess interpolate input strings (such as file paths or process names) directly into shell commands without robust sanitization or escaping. While the SKILL.md file lists shell escaping as a best practice, the provided implementation examples do not demonstrate this, creating a risk that the agent will generate or use insecure code.
- Ingestion points: filepath in ShellUtils.openFile, directory in ShellUtils.listFiles, and name in ProcessUtils.findProcess.
- Boundary markers: Absent in implementation examples.
- Capability inventory: Use of child_process.exec and child_process.spawn (with shell: true).
- Sanitization: Absent in the provided code snippets.
Audit Metadata