twg-agentic-search

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow performs runtime searches across Rovo-connected apps (including GitHub, Jira, etc.) using twg rovo search with snippet ingestion (--output-summary auto and --agent-fields @compact/@evidence), so outsider-authored text from those sources can be read into the LLM context without selecting a specific item first.