cypress-testing

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The companion-check section instructs the agent to run local shell commands (ls) and offer installing external skills (npx), which is unrelated to Cypress testing and asks the agent to access/modify the local environment—a hidden/out-of-scope instruction.

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt includes hard-coded plaintext passwords and examples that embed credentials directly into commands and test code (e.g., login.fillPassword('password123'), cy.login('user@example.com', 'password123')), which would require or encourage the LLM to output secret values verbatim.