csv-wave-pipeline

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs agents to read arbitrary project/reference files and to return and propagate "findings" and append discovery JSON verbatim (into prev_context, discoveries.ndjson, and report_agent_job_result), so any secrets present in those files would be output and propagated through the LLM and files—creating an exfiltration risk.