learning-guide-creator

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow explicitly requires conducting "10-15 web searches" and using web_fetch on the "5-8 best search results" (Phase 2 — Research), which means the agent will fetch and ingest open/public third-party web content and use it to drive decisions and guide writing, exposing it to untrusted user-generated or public sources that could contain indirect prompt-injection.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly requires runtime web_fetch of 5–8 external search results (i.e., arbitrary external URLs fetched at runtime via web_fetch) whose full article content is injected into the model context and thus directly influences/controls the agent's prompts and outputs.