pw-new-test-cli
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes user-provided test steps through the
$ARGUMENTSvariable, creating an indirect prompt injection surface. - Ingestion points: User-provided steps enter the agent's context in
SKILL.mdto guide test generation. - Boundary markers: Absent. There are no delimiters or warnings to ignore instructions embedded within the test steps.
- Capability inventory: The skill has extensive capabilities including writing files, executing shell commands (
npx playwright), making network requests via the Playwright browser, and performing Git commits. - Sanitization: Absent. The input is used as a direct logical guide for the agent's behavior.
- [COMMAND_EXECUTION]: The skill involves generating executable test scripts and running them using
npx playwright test. While this is the intended purpose, it represents the execution of dynamically generated code based on external input.
Audit Metadata