momentic-mobile-test
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data by inspecting mobile application UIs (via screenshots, native hierarchies, and webviews) and reading configuration files (
*.test.yaml,*.module.yaml). This presents a standard attack surface where malicious content within an application under test or a test file could attempt to influence the agent's behavior. - Ingestion points:
*.test.yamlfiles,*.module.yamlfiles, mobile device screenshots, native XML hierarchies, and webview state. - Boundary markers: The instructions do not specify explicit delimiters for external data, but standard agent guardrails apply.
- Capability inventory: The agent can write to the filesystem (editing YAML), execute shell commands (
npx momentic-mobile doctor), and run JavaScript. - Sanitization: No explicit sanitization of UI content or YAML fields is mentioned.
- [DYNAMIC_EXECUTION]: The skill instructions facilitate the execution of JavaScript for complex test logic, data generation, and assertions. It notes that v1 tests carry code inline while v2 tests use file-backed scripts. This is a core requirement for the skill's primary purpose of E2E testing.
- [COMMAND_EXECUTION]: The skill utilizes the
npx momentic-mobile doctorcommand to diagnose environment issues and connectivity. This is a vendor-provided CLI tool used for legitimate setup purposes. - [EXTERNAL_DOWNLOADS]: The skill references documentation and setup guides located at
momentic.ai. These are official vendor resources and do not represent a security risk.
Audit Metadata