From your agent to a business service.
What TeraBrain NX does, how to use it and how to keep it under control. For those who have already built an agent, for those who need to connect it to company systems and for those who must guarantee security and governance.
Getting started
From your agent to a service used by the whole company.
Bring an agent to production
The path from an agent prototype to a service your colleagues use, with access control, connections to your systems and oversight.
How to create a new agent
Step by step: project, agent node, instructions, model, tools, channel, testing and publishing.
Recipes: ready-made patterns
Six ready-wired starting projects: Telegram assistant, web app, mailbox, receptionist, agent with code, team of agents.
Build
Agents, channels, data, code and models: everything you can configure.
How it works: the graph project
Nodes, ports and links: the model used to describe an agent system, with versions and rollback.
Channels: how people reach agents
Web app, Telegram, mailbox, phone, scheduler and MCP server: how they connect and how you control who can use them.
Data and knowledge
MySQL databases, file storage and skill library: the persistent state of agents and code.
Multiple agents that collaborate
Delegating a task, handing over the conversation, splitting work into subtasks.
Telephony and voice
A small SIP switchboard in the project: lines, extensions, opening hours, menus and agents that talk on the phone.
LLM models
Providers and profiles: using cloud or private models, and choosing a different model for each agent.
Building with external assistants
Using Claude, Codex or other assistants to build and develop TeraBrain NX projects, with the person's own rights.
Integrate
Connect agents to systems, even those installed at the customer's site.
Security and governance
Who can do what, and how everything is tracked.