reverse-engineering-android-malware-with-jadx

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt explicitly instructs extracting and printing C2 URLs, IPs, base64-decoded strings and encryption keys from decompiled APKs (IOCs/secrets) so the agent would need to output those values verbatim, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.85). The workflow ingests the user-supplied APK by running tools like apktool d malware.apk and jadx ... malware.apk, which decompile outsider-authored malware code/resources into readable Java/manifest text that the LLM would then see (indirect prompt injection risk via attacker-controlled strings/code).