incident-ingesting

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This skill explicitly requires including "raw error messages, stack traces, or log excerpts" and attachments verbatim from Jira issues, which can contain API keys, tokens, or passwords and thus forces the LLM to output secrets if they are present in ticket content (even though the MCP handles auth for fetching).