ux-design-expert

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The template conditionally injects $SPICE_CLOUD_API_KEY into a command-line argument (--api-key), which requires the agent to handle and could output the API key verbatim (high exfiltration risk).

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

• Third-party content exposure detected (high risk: 0.85). The required workflow runs a hybrid RRF search over the ux_design_spec dataset and injects the retrieved content (document text authored outside the operating user) into the LLM context as “results” for synthesis, which is an outsider-authored free-text source.