hunt-subdomain

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill gives step-by-step, actionable instructions to discover and claim unclaimed subdomains — including registering/serving content (explicitly mentioning phishing and credential harvesters) and evasion techniques — which is potentially harmful dual-use guidance without clear benign-only constraint.

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

• Third-party content exposure detected (low risk: 0.10). SKILL.md describes a local/manual subdomain takeover hunting workflow (e.g., subfinder/amass/certificate transparency, then dig/curl/nuclei/subjack) that ingests only attacker-chosen target data and does not require the agent to read outsider-authored free text from third-party sources before selecting a specific item.