staging-subdomain-hunt

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). Most URLs are benign reconnaissance endpoints (crt.sh queries and the generic https://$DOMAIN) used for enumeration, but https://evil.com is an attacker/example domain used as a malicious Origin in the script and should be treated as suspicious and not trusted as a download source.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill explicitly instructs finding and exploiting staging/internal subdomains to perform site takeover, credential theft, and pivoting to production — clear malicious intent enabling data exfiltration and unauthorized access.

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

• Third-party content exposure detected (low risk: 0.10). SKILL.md is a bash workflow that fetches certificate transparency data from crt.sh (a provider endpoint) and then probes/queries attacker-selected targets (URLs/subdomains) via curl/httpx, so any outsider-authored free text ingested at runtime could come from the target’s HTTP responses or crt.sh output but the workflow does not include an internal “queue/feed” submission path that accepts arbitrary outsider text into the process without the operator selecting a specific target/subdomain.