routeros-scripting

Pass

Audited by Gen Agent Trust Hub on Nov 1, 2026

Risk Level: SAFE
Full Analysis
  • [DYNAMIC_EXECUTION]: The documentation describes the :parse command, which compiles strings into executable functions. This is presented as a standard language feature for dynamic command construction within the RouterOS environment.
  • [COMMAND_EXECUTION]: The skill covers the use of :execute, a built-in command for running scripts in the background or capturing output as a string. These are standard administrative functions for the platform.
  • [PRIVILEGE_ESCALATION]: The skill provides an overview of RouterOS script policies (permissions), such as sensitive, password, and policy. It correctly identifies these as critical for script correctness and safety, advising caution when using the dont-require-permissions flag.
  • [DATA_EXFILTRATION]: Mention of /tool/fetch is included as a method for retrieving external data or sending reports. The examples provided use generic URLs and are consistent with typical RouterOS automation tasks.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents how to ingest data from files or network sources. While it identifies these as ingestion points, it serves as a language reference rather than an implementation prone to injection, and it describes standard platform capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 1, 2026, 07:44 PM