lark-workflow-meeting-summary

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). In order to call CLI commands like drive metas batch_query and docs update the skill instructs the agent to record and embed note_doc_token/verbatim_doc_token values (resource tokens) verbatim into generated commands/requests, which requires the LLM to handle secret/resource tokens directly and creates an exfiltration risk.