mobile-recon-android

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs finding hardcoded secrets and shows reproduction/report steps that embed discovered credentials verbatim (e.g., grep patterns for API keys and export aws keys), so the agent would need to handle and output actual secret values.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The document contains explicit, actionable instructions for harvesting and abusing embedded credentials, exfiltrating data (AWS/Firebase/Stripe/Slack tokens), bypassing protections (SSL pinning, debuggable apps), and invoking exported components — behaviors that enable deliberate malicious activity.

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

• Third-party content exposure detected (high risk: 0.75). The required workflow ingests the target APK’s decoded text (e.g., AndroidManifest.xml, decompiled Java/smali) into the agent’s context via runtime file reads like cat apk-decompiled/AndroidManifest.xml and jadx output, where that APK content is outsider-authored (the app publisher/third party), creating an indirect prompt-injection risk.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill explicitly hunts for payment-gateway secrets (Stripe secret keys: greps for sk_live_/pk_live_), and includes examples that validate and use live credentials (AWS CLI commands to verify and enumerate access). It also shows triggering a PaymentActivity with an "amount" parameter. Finding and using Stripe secret keys enables direct financial actions (charging/refunding), so this skill contains specific functionality that can lead to direct financial execution authority.