gradle-patterns

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No evidence of instructions attempting to override agent behavior or safety guidelines was found. The skill contains purely technical documentation and Gradle configuration templates.
  • [EXTERNAL_DOWNLOADS]: The skill references standard Android and Kotlin dependencies from official repositories. All modules (e.g., Google's AGP, AndroidX Compose, JetBrains Kotlin, Ktor, and Koin) are well-known, reputable libraries for the Android ecosystem.
  • [COMMAND_EXECUTION]: The skill lists common Gradle CLI commands for building, profiling, and cache maintenance (e.g., ./gradlew build, rm -rf ~/.gradle/caches). These are standard development workflows and do not involve arbitrary or malicious execution.
  • [DATA_EXFILTRATION]: No sensitive file paths, credential exfiltration patterns, or network operations to non-standard domains were detected. The use of standard Gradle repositories (Google, Maven Central) is normal and safe.
  • [OBFUSCATION]: No obfuscated strings, hidden characters, or encoding techniques designed to conceal intent were found in the skill content.
  • [DEPENDENCIES]: The version catalog (libs.versions.toml) defines standard, versioned dependencies for Android projects. No suspicious or typosquatted packages were observed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 01:14 PM
Security Audit — agent-trust-hub — gradle-patterns