ai-research-explore

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

• Third-party content exposure detected (high risk: 0.85). Yes: the workflow performs free-first research lookup that fetches external web content (e.g., via scripts/lookup/providers/url_provider.py calling http_get and parsing page text into summary/title) and then feeds the resulting lookup records into the LLM context via scripts/orchestrate_explore.py’s build_context() (under lookup_records/sources_* fields).