android-apk-reverse-engineering
Fail
Audited by Snyk on Jul 22, 2026
Risk Level: CRITICAL
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 1.00). The prompt contains literal hardcoded-looking credentials (AKIA... and an AWS secret) and explicitly instructs using discovered secrets to exploit infrastructure, which requires the LLM to handle/output secret values verbatim.
CRITICAL E006: Malicious code pattern detected in skill scripts.
- Malicious code pattern detected (high risk: 1.00). The skill documentation contains explicit, actionable guidance to locate hardcoded credentials, bypass SSL pinning, recompile/sign/install modified APKs, and even a flowchart step to "use secrets to exploit backend infrastructure," which enables credential theft, data exfiltration, and repackaging/supply-chain abuse.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.75). The required workflow in SKILL.md instructs the agent to download an APK from a third-party mirroring site (e.g., APKMirror) and then to decode/decompile that APK into human-readable XML/smali/Java via tools (apktool/jadx), so outsider-authored content from a public web source is likely ingested into the LLM context during analysis.
HIGH W008: Secret detected in skill content (API keys, tokens, passwords).
- Secret detected (high risk: 1.00). The SKILL.md contains literal, non-placeholder credentials. The string "AKIA1234567890ABCDEF" matches the AWS access key format (starts with "AKIA" and is 20 chars) and is a hardcoded API key in the Java example. The accompanying AWS_SECRET value ("ZxyYxWwVvUuTtSsRrQqPpOoNnMmLlKkJjIiHhGgFf") is a long, specific secret-like string (not a simple documentation placeholder) and is presented as the AWS secret in the same code block. Both are concrete, high-entropy-looking values and should be treated as real credentials rather than placeholders.
Issues (4)
W007
HIGHInsecure credential handling detected in skill instructions.
E006
CRITICALMalicious code pattern detected in skill scripts.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W008
HIGHSecret detected in skill content (API keys, tokens, passwords).
Audit Metadata