playwright-reviewing
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill suggests using standard Unix utilities (
find,grep,ls) to search local files for specific code patterns. These are read-only operations that do not execute the files or modify the system. - [DATA_EXFILTRATION] (SAFE): No network operations, credential access, or data transmission commands were detected.
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill is designed to search through untrusted user-provided test files. While these files could contain malicious instructions, the risk is limited because the skill uses static pattern matching (
grep) rather than interpreting or executing the file contents. - Ingestion points: Local
.spec.tsand.test.tsfiles analyzed via bash commands. - Boundary markers: Absent.
- Capability inventory: Shell command execution for searching and listing files.
- Sanitization: Absent.
Audit Metadata