The Model Context Protocol (MCP) orchestrates AI tools, security context, and first-party/third-party data. It powers automated workflows and explainable outputs across the platform.
ATTOM data (valuations, comps, tax), model registry (OpenAI, Azure OpenAI, Anthropic), and identity/RBAC (SSO + SCIM, audit context) form the inputs MCP uses per request.
Model Registry
Models are selected based on task and policy. The registry allows easy addition of providers and variants, wired to user-facing AI settings.
Security Context
Every call carries workspace/user identity, roles, and correlationId. Approvals are enforced before sensitive actions. All events are auditable.
Events & Streaming
Real-time updates stream over Server-Sent Events (SSE). Standardized ai.* events let the UI update feeds, show statuses, and correlate actions across surfaces.
Approvals & Audit
Approval states are persisted (TypeORM + Postgres) with actor attribution, reasons, and correlationId. SSE broadcasts keep the UI consistent everywhere.
AI Insights
Outputs include explainability and source tags (e.g., ATTOM vs. simulated). This keeps compliance teams confident and speeds operator decisions.
Feature Tie-ins
Calendar
AI schedules follow-ups and SLA alerts using workflow triggers.Go to Calendar
Document Center
Generate, QA, and redline with audit-friendly explanations.Go to Documents
Tasks
Tasks are created from chat and approvals; correlated end-to-end.Go to Tasks
Collaboration
Mentions and summaries propagate via ai.* events to keep teams aligned.Go to Collaboration
Deal Pipeline
ATTOM-backed valuations and risk scores inform stage movement.Go to Pipeline
Automation
workflow.trigger runs gated automations with full auditing.Go to Automation
OrbiLattice – AI Property Signals to Orchestrated Action