incident-activity-summarization

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill explicitly instructs the agent to fetch and merge journal entries, work notes, and comments into generated timelines and summaries (and to save those summaries as work notes), which means any secret strings present in those fetched fields could be reproduced verbatim in the LLM output and thus exfiltrated.