routeros-scripting
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze RouterOS script files (.rsc), which represent an untrusted data surface.
- Ingestion points: Content of .rsc files processed by scripts/dry-run-import.sh and scripts/dry-run-import.ps1.
- Boundary markers: The skill does not implement specific delimiters or 'ignore' instructions to prevent the agent from obeying instructions embedded within the processed scripts.
- Capability inventory: The skill can execute remote commands (ssh), upload files (scp), and modify network configurations.
- Sanitization: No validation or sanitization of the .rsc file content is performed prior to processing.
- [COMMAND_EXECUTION]: The skill includes shell and PowerShell scripts that invoke system binaries (ssh, scp) to manage remote routers. While these tools are required for the skill's primary function, they create a surface for command execution if the agent is directed to use malicious parameters for the host, user, or file path fields.
Audit Metadata