copywriting

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The prompt embeds mandatory developer-oriented commands (e.g., "search codebase for 3+ similar patterns", "cite file:line evidence", and break work into TaskCreate including per-file reads) that are unrelated to the copywriting purpose and attempt to compel file access/exfiltration, so it functions as a prompt injection outside the skill's stated scope.

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt mandates reading project files and citing file:line evidence for every claim, which can force the agent to reproduce file contents (potentially including API keys, tokens, or passwords) verbatim and thus creates an exfiltration risk.