tutorial-creator

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

• Third-party content exposure detected (high risk: 0.85). Yes: the required “External source” workflow (Path 1 entry [f] / Path 2 entry [d]) accepts a URL or pasted text from the user, then fetches/reads that outsider-authored free text at runtime and uses it as input for generation in the LLM context (indirect prompt-injection risk).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches user-provided URLs at runtime (see SKILL.md/AUDIENCE.md entry for External source) and uses the fetched page content to extract concepts and drive tutorial generation (which directly controls model prompts); an example external URL present in the repo is https://stuffolio.app.