home-assistant-card

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md requires fetching Home Assistant developer documentation from a public GitHub URL at runtime (public web content), which is outsider-authored free text that can be ingested into the agent’s LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs the agent at runtime to fetch authoritative markdown from https://raw.githubusercontent.com/home-assistant/developers.home-assistant/master/docs/, which will be used to guide and control implementation decisions and prompts (a required external dependency fetched during runtime).