content-reviewer

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

• Third-party content exposure detected (high risk: 0.85). The required workflow ingests the influencer’s submission text (link/description/caption/media notes) provided by an outsider creator, which the LLM then uses during the compliance/brand/message review steps (SKILL.md:31-32, 68-75).

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's standalone-install fallback (line 64) explicitly instructs fetching the runbook at runtime from https://raw.githubusercontent.com/aaron-he-zhu/aaron-marketing-skills/main/references/, which is a required dependency that directly controls prompts/scoring ("Do not score without the runbook").