skill

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill asks the user to provide their VISION_AGENT_API_KEY (unless they say "env") and explicitly instructs the agent to store and embed that value directly into generated curl commands/Authorization headers, which requires outputting the secret verbatim.

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

• Third-party content exposure detected (high risk: 0.90). The skill accepts arbitrary document URLs (Step 2: "Local file path or URL?" with document_url and quick-reference markdown_url) and explicitly instructs the agent to read parsed markdown (Schema Builder → "Read the parsed markdown content") to generate extraction schemas and drive subsequent extract/split operations, so untrusted third-party content can materially influence tool use and next actions.