demo-builder

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly asks the user to provide "Any API keys or credentials I should use" and directs the agent to read authentication details and produce example API/CLI usage (e.g., curl), which requires embedding secrets verbatim and creates an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs the agent to “Research the selected prospect company thoroughly” using web search and the company’s website/news/LinkedIn, which would require fetching and ingesting outsider-authored public web text into the LLM context at runtime.