security-investigator-automation

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). The links are primarily to a GitHub repository from an unvetted/unknown user plus custom domains (custom-ti-api.com, ara.so) referenced by the project—while YouTube is benign, the repo and external API endpoints represent a supply‑chain risk if you clone/run code or call those endpoints without code review and trusted provenance.

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow performs threat-intel enrichment by calling external IP reputation/lookup services (e.g., AbuseIPDB, Shodan, IPInfo, VPNAPI) and then injects the returned enrichment text into the agent’s LLM context for analysis/reporting; these are outsider-authored free-form responses from public third-party APIs.