routeros-troubleshooting
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides Bash and PowerShell scripts (
collect-diagnostics.sh,collect-diagnostics.ps1) to automate data collection from remote routers via SSH. - The scripts are designed to use SSH keys or agents (BatchMode), avoiding the need to handle or store passwords.
- Executed commands are limited to a hardcoded list of read-only diagnostic prints (e.g.,
/system/resource/print,/log/print). - The scripts include filename sanitization to prevent path traversal when saving output files.
- [INDIRECT_PROMPT_INJECTION]: The diagnostic collection process ingests data from external network devices, which is a known attack surface for indirect prompt injection.
- Ingestion points: Remote command output retrieved via SSH in
scripts/collect-diagnostics.shandscripts/collect-diagnostics.ps1. - Boundary markers: None; the scripts collect raw text output and save it to a file.
- Capability inventory: The skill has the ability to write files to the local system and execute SSH commands for data retrieval.
- Sanitization: No sanitization is performed on the ingested log lines or device output, as they are intended for verbatim diagnostic review.
Audit Metadata