100x-persona

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.95). The skill mandates echoing verbatim substrings from the user-provided source_script into output fields (e.g., evidence_quote / script_span_quote), so any secrets included in the source_script would be reproduced verbatim by the LLM, creating an exfiltration risk.