Voice Platform for Business
VOX (Latin for Voice or Expression) is a multi-tenant voice agent platform for building production-ready voice experiences — from product sales and fulfillment to reservations and customer support — grounded in your real business data. You define how the agent behaves (prompt + dialog flow), what it can say (your data as context), what it can do (tools), and what it can show (UI components). The platform handles the runtime: session creation, safety controls, tool execution, and consistent rendering of results.
Documents
Plan the agent’s purpose, tone, workflows, and success criteria before you build.
Author the system prompt: tool rules, dialog flow, guardrails, templates, and examples.
Define HTTP tools that connect the agent to your APIs and data with reliable UI outcomes.
Connect tenants, agents, and widget keys. Manage identity, limits, and deployment settings.
Core Business Features
| Business Capability | What it delivers | Business impact |
|---|---|---|
| 24×7 Customer Coverage | Voice agents are always available to answer questions, take requests, and complete transactions without staffing constraints. | Improves customer satisfaction while reducing after-hours staffing costs and missed opportunities. |
| Higher Conversion and Sales Enablement | Agents actively guide customers through product discovery, availability, and booking or purchase flows. | Increases conversion rates and average order value by assisting customers at the moment of intent. |
| Faster Customer Resolution | Agents instantly retrieve accurate information about orders, inventory, reservations, and policies. | Reduces wait times, lowers support tickets, and improves first-contact resolution. |
| Consistent Brand Representation | Agents follow defined tone, messaging, and behavior aligned to your brand voice and service standards. | Protects brand reputation and ensures uniform customer experiences across channels. |
| Scalable Customer Engagement | One platform supports thousands of simultaneous conversations without service degradation. | Enables growth without proportional increases in headcount or infrastructure cost. |
| Employee Productivity Assistants | Internal agents help staff find answers, search products, check availability, and complete routine tasks. | Frees employees to focus on higher-value work and reduces onboarding and training time. |
| Revenue-Generating Service Automation | Agents can complete bookings, schedule appointments, process payments, and capture qualified leads. | Turns customer service into a profit center instead of a cost center. |
| Personalized, Context-Aware Interactions | Agents tailor responses using customer context, history, and real-time business data. | Creates more relevant conversations that build trust and improve loyalty. |
| Rapid Time-to-Value | New agents, workflows, and offers can be launched and updated quickly without major system changes. | Allows marketing and operations teams to respond faster to market opportunities. |
| Enterprise-grade Reliability and Cost Controls | Built-in safeguards manage usage, prevent abuse, and keep spending within defined limits. | Protects margins and ensures predictable operating costs as usage scales. |
| Monitored | All conversations are monitored, and escalated for 'human in the loop' intervention where required. | Excellent customer care and responsiveness. |
| Secure | With our powerful tool sets, data and configurations are managed through your set of APIs | Your data. Protected. |
Core Technical Features
| Feature | What it provides | Why it matters in production |
|---|---|---|
| Data-grounded conversations | Agents retrieve real-time, structured JSON from your systems via tools (availability, inventory, orders, events, pricing). | Prevents hallucinations and ensures answers reflect current business reality. |
| HTTP Tool Descriptors (schema-driven) | Declarative JSON schema for methods, parameters, auth headers, URL/body templates, and timeouts. | Standardizes every integration and avoids one-off runtime code per API. |
| Declarative UI rendering | Tools can declare UI actions on success/error (catalogs, confirmations, payment forms, media galleries). | Turns API responses into consistent customer experiences automatically. |
| Built-in core tools | Platform utilities like time checks, website scraping, opening links, clipboard copy, and component rendering. | Accelerates your Agent design and configuration without custom setup for common tasks. |
| Prompt-driven workflows | System prompt defines tone, guardrails, tool triggers, dialog flow, and response templates. | Keeps behavior predictable and auditable across updates and versions. |
| Strong templating + typing rules | Stable execution context (args, response, status, secrets) with filters for numbers, booleans, currency, and JSON objects. | Prevents integration bugs such as stringified numbers or broken UI props. |
| Multi-tenant configuration model | Tenants, agents, widget keys, origins, limits, and API secrets are centrally managed. | Enables secure isolation and scalable onboarding of many customers. |
| Layered security & abuse protection | Edge + server rate limits, bot verification, session quotas, duration and idle caps, daily spend limits. | Protects reliability and cost without disrupting real users. |
| Session observability & lifecycle control | Realtime session records, heartbeats, stale detection, and mid-session termination. | Keeps dashboards accurate and prevents runaway sessions. |
| Fast test-and-iterate loop | Start agents instantly, inspect tool calls, validate APIs, and refine prompts. | Shortens feedback cycles and de-risks production launches. |
How the platform fits together
- Agent — personality, guardrails, and workflow (prompt)
- Tools — what the agent can do (HTTP descriptors calling your APIs)
- Data — what the agent knows (structured JSON returned by your systems)
- UI Components — what the agent can show (forms, confirmations, galleries, results)
- Tenant Configuration — who can use it and where (tenantId, widget keys, origins, limits)
This model keeps your voice experience grounded, secure, and maintainable—while staying flexible enough for bookings, commerce, scheduling, and support across different businesses.
Before you dive into building your first agent, we recommend trying out our demo agents at strategicmachines.ai. This will give you a hands-on feel for how voice agents work in real scenarios, such as reservations or product searches. Once you're familiar, follow the sections below to design and configure your own.