chat-summarization-va

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly retrieves context entries including the "value" field and instructs the agent to include captured slot values verbatim in generated summaries/work_notes (and shows examples with slot values), which requires the LLM to handle/output sensitive values directly.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). Step 2 (and 1) uses MCP/REST to query and retrieve full chat transcripts (sys_cs_message.body, formatted_body, and related context/slots) from conversation records that contain outsider-authored user messages.