seo-email

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The preflight step explicitly instructs the agent to stop, display a setup wizard, solicit the user's DataForSEO login/password, parse and write them to disk — behavior that requests sensitive credentials and contradicts the skill's stated "No new API calls / purely a drafting step", so it's a hidden/deceptive instruction outside the skill's advertised purpose.

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill tells the agent to ask the user to paste their login and password, parse those values, and write them verbatim into a .env file (and use them for verification), which requires the LLM to handle and output secrets directly — a high exfiltration risk.