frontend-testing-debugging
Pass
Audited by Gen Agent Trust Hub on May 9, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- Web Content Processing (Indirect Surface): The skill is designed to interact with and analyze external web content, including DOM snapshots, page titles, and console logs, to identify and fix UI bugs. This is a common pattern for browser automation tools.
- Ingestion points: Browser data is collected via
tab.playwright.domSnapshot()andtab.dev.logs()inSKILL.md. - Boundary markers: The skill does not explicitly define delimiters for untrusted web content.
- Capability inventory: The agent has the ability to execute shell commands (via package managers) and modify local source code to address findings.
- Sanitization: Web content is analyzed directly to verify application state and console health.
- Dynamic Script Generation: To perform validation when the primary Browser plugin is unavailable, the skill generates and executes temporary Playwright scripts. These scripts are used for localized debugging and are kept outside the repository's committed source code.
- Local Command Execution: The skill utilizes standard project scripts defined in
package.jsonand uses tools likepnpm,npm, oryarnto run tests and capture screenshots. This interaction is scoped to the local development environment.
Audit Metadata