Fact Check

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow uses WebSearch/WebFetch to retrieve and extract external web pages/specifications (outsider-authored public content) into the agent’s LLM context for verification.

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

• Potentially malicious external URL detected (high risk: 0.75). The skill's WebFetch tool fetches arbitrary URLs at runtime (e.g., developer.mozilla.org or raw GitHub URLs such as https://raw.githubusercontent.com/...), and that fetched content is injected into the agent's verification context to drive prompts/instructions, so it is a runtime external dependency that can directly control the agent.