research

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly uses third-party web tools—e.g., firecrawl.search/scrape (for tutorials, Stack Overflow, blogs) and octocode.githubSearchCode/githubSearchIssues (for public GitHub repos/issues)—and requires the agent to ingest and synthesize that untrusted, user-generated web content as part of its workflow, exposing it to indirect prompt injection.