ai-writing-auditor

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt requires the agent to cite and return "exact text" from the target content and to output the full rewritten content and findings table verbatim, so if the input contains secrets (API keys, tokens, passwords) the agent would be forced to reproduce them in its output, creating an exfiltration risk.