meta-ad-policy-checker

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text is fetched at runtime from public Meta Transparency Center pages (e.g., the ad-standards index and resolved policy pages) and then ingested into the agent’s LLM context for citation/reasoning in Phase 2 (“Fetch + Cache Live Policy Text”).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches live policy pages at runtime (starting with https://transparency.meta.com/policies/ad-standards/ and then resolves and fetches canonical Meta policy pages) and injects that fetched policy text into the agent's reasoning/output, so external content directly controls the agent's prompts/instructions and is a required dependency.