agency-threat-detection-engineer

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly instructs the agent to "Review threat intelligence feeds, industry reports, and MITRE ATT&CK updates" and to "build automated pipelines that ingest IOCs from STIX/TAXII feeds," which clearly requires fetching and interpreting external third‑party threat‑intelligence content that could influence detection decisions.

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

• Potentially malicious external URL detected (high risk: 0.70). The GitHub Actions workflow includes runtime steps that fetch and execute remote code (e.g., uses: actions/checkout@v4 and pip install sigma-cli / backends) and deploys via curl to https://${{ secrets.SPLUNK_HOST }}:8089/servicesNS/..., so external content is fetched and executed as required dependencies during CI/runtime.