qa-ui-test
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The
scripts/setup.jsscript usesexecSyncto initialize the project, install the Playwright framework, and provision browser binaries. - [EXTERNAL_DOWNLOADS]: The skill fetches necessary testing dependencies and browser binaries from official registries during the initialization phase.
- [REMOTE_CODE_EXECUTION]: The agent creates and runs executable Playwright test scripts (
.spec.ts) based on the layout and content of analyzed web pages. - [PROMPT_INJECTION]: The process of analyzing external web content to generate scenarios poses a risk of indirect prompt injection.
- Ingestion points: Untrusted web page content read via Playwright navigation in
SKILL.mdStep 1. - Boundary markers: Not explicitly implemented to isolate external data from internal agent instructions.
- Capability inventory: Writing files (test scripts, data) and executing shell commands (Playwright test runner).
- Sanitization: No specific mechanisms for sanitizing or escaping ingested web content are defined in the instructions.
Audit Metadata