xml-to-compose-migration
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where an AI agent ingests and processes untrusted Android XML layout data to generate Jetpack Compose code. This creates an attack surface where malicious instructions could be embedded within the XML (e.g., in strings, comments, or custom attributes) to influence the agent's behavior or output code.
- Ingestion points: The agent reads user-provided XML layouts and View widget attributes to perform mapping (SKILL.md).
- Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions to isolate the untrusted XML content from the agent's core migration logic.
- Capability inventory: The skill guides the agent to generate Kotlin code, migrate application state, and potentially integrate navigation, providing a path for injected instructions to manifest in executable code.
- Sanitization: There are no instructions for validating or sanitizing the input XML to ensure it contains only layout definitions and not adversarial text.
Audit Metadata