testing-e2e
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to run tests vianpx playwright test, install dependencies withnpm install, and execute helper scripts vianode. - [EXTERNAL_DOWNLOADS]: Automatically fetches the
@playwright/testpackage and browser binaries from the official NPM registry and Microsoft servers, which are trusted sources. - [REMOTE_CODE_EXECUTION]: Executes Playwright test scripts dynamically generated by the agent and stored in the
/tmpdirectory, which is a core function for automated testing tasks. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it ingests local project files to inform its testing logic.
- Ingestion points: Reads project files including
package.json,playwright.config.ts, and**/*.spec.tsfiles. - Boundary markers: Does not implement delimiters or safety instructions when reading the contents of these files.
- Capability inventory: Has access to
Bashfor shell command execution andTaskfor spawning sub-agents. - Sanitization: Content from the ingested files is used without explicit sanitization to determine execution parameters like dev server commands.
Audit Metadata