social-captions

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly accepts and analyzes competitor posts and arbitrary links from the open web (see README/Usage examples: "competitor post ... [url or text]" and "A link to an article you're repurposing") so the agent ingests untrusted public third‑party content and uses it to drive caption-generation decisions, creating a risk of indirect prompt injection.