longbridge-value-screen

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

• Third-party content exposure detected (high risk: 0.75). The required workflow fetches constituent lists and valuation/dividend data via the longbridge CLI/MCP at runtime (e.g., longbridge constituent <INDEX> --format json and related calls), which is outsider-provided free text/JSON that the agent will read and include in its LLM context.