play-store-release
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from local project files and maintaining access to shell execution tools. \n
- Ingestion points: The skill processes files like
app/build.gradle.kts,proguard-rules.pro, and CI configuration files found in the target repository. \n - Boundary markers: No explicit markers or instructions are provided to the agent to treat data in these files as non-executable text or to ignore embedded instructions. \n
- Capability inventory: The skill defines capabilities for executing shell commands via Gradle and editing repository files. \n
- Sanitization: There is no specified logic for sanitizing or validating the content extracted from the project files before it enters the agent context. \n- [COMMAND_EXECUTION]: Provides valid instructions for executing standard Android build and linting commands, such as
./gradlew bundleReleaseand./gradlew lint. \n- [EXTERNAL_DOWNLOADS]: References documentation and installation patterns from Vercel Labs' public repository for maintaining agent skills.
Audit Metadata