brand-kit

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

• Third-party content exposure detected (high risk: 0.95). The skill explicitly ingests open web and user-generated content as required by its workflow — e.g., Step 0 instructs scraping URLs and fetching social profiles with FirecrawlScrapeTool, FetchInstagramProfileTool, FetchTiktokProfileTool and FacebookBusinessPageInfoTool, Step 1 uses PerplexitySonarSearchTool/GoogleNewsSearchTool and FirecrawlScrapeTool to research competitor websites and social profiles, and Steps 6–7 use WebFetch to check domains and platform profile URLs — all of which the agent is expected to read and which materially influence strategy and subsequent actions.