ioc-extract

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill reads arbitrary text/log files/URLs and writes/prints extracted IOCs (including full URLs and emails) verbatim to files and terminal, so any API keys, tokens, or passwords present in the input would be output unchanged and could be exfiltrated.

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

• Third-party content exposure detected (high risk: 0.80). The skill's Step 1 workflow explicitly fetches arbitrary URLs via curl when INPUT_SOURCE is an http(s) URL, ingesting untrusted public web content that is parsed for IOCs and whose results the skill (and its notes) can drive follow-up tool use (e.g., feeding hashes into the malware-hash/VirusTotal flow), so third-party content can materially influence subsequent actions.