premortem

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt tells the agent to scan workspace files and include "full context" from those files in sub-agent prompts, the HTML report, and the transcript — which can cause any secrets present in workspace files (API keys, tokens, passwords) to be copied verbatim into outputs, creating an exfiltration risk.