synthesize-documents

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs the agent to "preserve the full content surface" and include concrete examples, evidence, and source-extracted points from provided documents/attachments, which can force verbatim reproduction of any API keys, tokens, or passwords present in those inputs and therefore risks secret exfiltration.

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly accepts URLs as inputs (Step 1: "including ... URLs") and mandates using external search and citing links when verifying claims (Step 6: "When using external sources... cite links"), so the agent will fetch and interpret open/public third‑party content that can materially influence its synthesis and actions.