compose-theming-styles
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core workflow involves analyzing untrusted local project files, including theme definitions and UI components, which serves as a potential vector for indirect prompt injection.
- Ingestion points: The agent is instructed to locate and analyze
Theme.kt,build.gradle.kts, and custom UI component files (Step 1 and Step 3). - Boundary markers: The instructions do not define clear boundary markers or provide 'ignore embedded instructions' warnings for the data being read from the project files.
- Capability inventory: The skill possesses the capability to write new files (e.g.,
ComponentStyles.kt), modify existing source code, and update build scripts (build.gradle.kts). - Sanitization: There is no mention of sanitizing or validating project content before it is processed by the agent or incorporated into new code generation.
- [COMMAND_EXECUTION]: The migration workflow (Step 3.1) directs the agent to execute or create and run screenshot tests using UI Automator or Espresso. This involves executing code and interacting with the local development environment (emulator or device), representing a command execution surface common in development-focused skills.
Audit Metadata