network-config-validation

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's checks both detect password/SNMP community patterns and the dangerous-command routine returns the matched command text verbatim (e.g., "command": stripped), which would cause secrets present in pasted configs to be included in output.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The skill’s workflow ingests configuration snippets provided by the user/automation (“review … snippets” and “Run … checks on the exact snippet to be pasted” / “Run … checks against the full candidate config”), so outsider-authored free text in the candidate config is directly processed by regex/IP parsing at runtime.