ot-security-assessment

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes an explicit curl example with a header "apiKey: " and instructs replacing placeholders with real values, which requires embedding a secret verbatim into generated commands/outputs (high exfiltration risk).

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

• Third-party content exposure detected (high risk: 0.85). The workflow’s “Online Vulnerability Research” step uses curl to fetch free-form web content from outsider-authored sources (e.g., NVD/CISA/MITRE pages) and that fetched text/JSON is then ingested into the agent’s runtime context for vulnerability analysis.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill includes a runtime curl command that downloads and installs an Nmap NSE script from https://raw.githubusercontent.com/nmap/nmap/master/scripts/dnp3-info.nse which fetches remote code that will be executed by nmap for DNP3 enumeration, and the skill relies on that fetched script as a required dependency.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill explicitly instructs use of sudo (e.g., sudo nmap -sU), modifying capabilities via sudo setcap, and writing to system directories (curl to /usr/local/share/...), which direct the agent to perform privileged actions that modify the host system state.