halo-asset-context

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

• Third-party content exposure detected (medium risk: 0.30). In mcp-servers/halo-mcp/halo_mcp_server.py, the workflow can ingest outsider-provided free-text via the halo_list_assets(search=...) parameter to search “Devices” (assets) and then read related tickets/relationships for the matched asset(s), creating an indirect prompt-injection surface through Halo-held ticket/comment text.