cypress-skill
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or security risks were detected in the skill instructions or provided examples.
- [COMMAND_EXECUTION]: The skill references standard execution commands for the Cypress framework (
npx cypress) and the LambdaTest cloud platform (npx lambdatest-cypress). These tools are well-known in the QA automation industry and used as intended for the skill's primary purpose. - [CREDENTIALS_UNSAFE]: The skill correctly instructs users to manage authentication credentials using environment variables (e.g.,
${LT_USERNAME}) rather than hardcoding sensitive data. It also demonstrates the use of{ log: false }to prevent secret exposure in Cypress test logs. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-supplied requirements and page URLs to generate test code. While this introduces a potential injection surface for untrusted data, the skill does not possess the combined capabilities (such as automated network exfiltration of local sensitive files) required to exploit this surface.
Audit Metadata