brightdata

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

• Third-party content exposure detected (high risk: 0.95). The workflow’s Tier 1/2/3/4 URL-content retrieval path ingests outsider-authored free text from public web pages at runtime (user-supplied arbitrary URLs fetched via WebFetch/curl/Playwright/Bright Data MCP), which is then converted to markdown and fed into the agent’s LLM context.