mobile-ux-ci
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill generates and suggests the execution of standard Playwright commands (e.g.,
npx playwright test) and CI workflow steps (e.g.,npm ci,npx playwright install). These are legitimate actions for the skill's stated purpose of automated testing. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting data from a sub-agent's exploration of the codebase.
- Ingestion points: Data regarding app routes, structure, and interactive elements enter the context via the 'Explore' agent output in Phase 2.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when the agent processes the sub-agent's findings.
- Capability inventory: The skill has the capability to write files (test specifications and CI workflows) and execute commands via the agent's CLI tools.
- Sanitization: There is no explicit sanitization or validation logic described for the data returned by the sub-agent before it is interpolated into generated test files.
Audit Metadata