e2e-qa-tester

Warn

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONCREDENTIALS_UNSAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes PowerShell via Get-ChildItem and Test-NetConnection to recursively search the project directory for credential files and verify local port availability.\n- [CREDENTIALS_UNSAFE]: The file references/test-patterns.md contains hardcoded plaintext credentials (e.g., admin123) used for login examples in the documentation.\n- [DATA_EXFILTRATION]: The workflow involves an automated recursive search for CREDENTIALS.md, a sensitive file path typically containing secrets, which could lead to unauthorized data exposure.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. \n
  • Ingestion points: Processes conversation history to determine testing tasks and interacts with web application content via Playwright.\n
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the processing logic.\n
  • Capability inventory: Full access to Playwright MCP tools (navigation, clicking, typing) and PowerShell command execution.\n
  • Sanitization: No sanitization or validation of the identified tasks or web content is performed before execution.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 25, 2026, 03:48 AM