cometchat-react-testing

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation and code templates for setting up testing environments (Vitest, Playwright). All code examples follow standard software engineering practices for testing.
  • [CREDENTIALS_UNSAFE]: The skill explicitly includes a meta-test pattern designed to prevent hardcoded credentials (Auth Keys) in the source code. It correctly identifies the use of environment variables and CI secrets as the secure method for handling API keys.
  • [EXTERNAL_DOWNLOADS]: The skill references official documentation and well-known testing tools (Vitest, Playwright, React Testing Library). These are well-known services and do not pose a security risk.
  • [COMMAND_EXECUTION]: The skill contains standard installation commands (npm install) and test execution commands (npm run test). These are routine for a development-focused skill and are not used for malicious purposes.
  • [DATA_EXFILTRATION]: There is no evidence of data exfiltration. The network operations described are limited to local testing (e.g., localhost) or standard WebSocket behavior within the CometChat SDK's intended scope.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 11:41 AM
Security Audit — agent-trust-hub — cometchat-react-testing