test
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches Playwright dependencies and browser drivers from official registries to support browser automation tasks.\n- [COMMAND_EXECUTION]: Executes shell commands for environment setup, test execution, and comprehensive cleanup of temporary logs and artifacts.\n- [REMOTE_CODE_EXECUTION]: Generates and runs TypeScript test scripts dynamically in CI mode to automate verification steps derived from project documentation.\n- [DATA_EXFILTRATION]: Processes implementation details and acceptance criteria from external task documents to drive browser navigation and interaction, representing an indirect prompt injection surface where external inputs influence agent-driven web actions.
Audit Metadata