android-native-dev

Pass

Audited by Gen Agent Trust Hub on Mar 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes standard developer commands for building Android applications using the Gradle wrapper (e.g., ./gradlew assembleDebug) and interacting with connected devices or emulators through the Android Debug Bridge (e.g., adb shell am start). These are routine operations for Android development.
  • [EXTERNAL_DOWNLOADS]: The skill references well-known and trusted Android libraries from the official Google and AndroidX ecosystems (e.g., androidx.compose:compose-bom). These resources are from established vendors and are considered safe for development purposes.
  • [PROMPT_INJECTION]: The skill presents a potential indirect prompt injection surface because it instructs the agent to analyze and complete user-provided project files while maintaining shell execution capabilities (Gradle and ADB). 1. Ingestion points: Project scenario assessment and file checklist in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Shell execution via ./gradlew and adb commands. 4. Sanitization: Absent. This vulnerability surface is inherent to the primary purpose of a developer-oriented skill and is mitigated by the skill's own emphasis on security practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 26, 2026, 08:07 AM