luma-event-attendees

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

• Third-party content exposure detected (high risk: 0.80). This skill directly fetches and parses public Luma event pages and Apify-scraped Luma results (see scripts/apify_client.py LumaClient.scrape_event and LumaApifyClient._run_actor, and SKILL.md "AI Agent Workflow"), ingesting untrusted, user-generated guest/profile bios and links which the agent is then expected to read and use to qualify leads and generate outreach, so those third-party contents could materially influence agent actions.