firebase-apk-scanner
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- COMMAND_EXECUTION (LOW): The skill utilizes Bash to execute a local script (scanner.sh) and the apktool utility for reverse engineering Android applications. This is standard behavior for a security auditing tool and is restricted to the provided arguments.\n- EXTERNAL_DOWNLOADS (LOW): The skill uses curl to interact with Firebase and Google API endpoints (identitytoolkit.googleapis.com, firebaseio.com, etc.). Because these target trusted organizations (Google/googleapis) for the skill's primary purpose, the severity is downgraded per security guidelines.\n- PROMPT_INJECTION (LOW): Vulnerable to indirect prompt injection (Category 8). Evidence: 1. Ingestion points: APK files ($ARGUMENTS) and extracted strings/resources. 2. Boundary markers: Absent in the workflow when presenting scan_report.txt results to the agent. 3. Capability inventory: Bash command execution and curl network access. 4. Sanitization: No evidence of output sanitization to prevent malicious APK content from influencing the agent's logic.
Audit Metadata