AI Agents
Early Alpha Version
AI Agents are in early alpha. Node interfaces, behaviors, and APIs are subject to change.
Overview
XGENIA AI Agents let you build AI-powered features visually — no backend code needed. Wire together a few nodes and you have a working chatbot, content generator, or intelligent assistant running directly inside your app.
The system is built around three core ideas:
- Visual-first: Define agents, tools, and workflows by placing and connecting nodes — just like the rest of XGENIA.
- Provider-agnostic: Switch between OpenRouter, OpenAI, Anthropic, Groq, and other LLM providers with a dropdown.
- Auto-discovery: Drop Tool nodes into the same component as an Agent and they wire themselves up automatically.
Architecture
The basic agent setup uses three nodes:
[LLM Provider] → [Agent] → [Agent Chat]
| Node | Role |
|---|---|
| LLM Provider | Picks the AI model and provider (OpenRouter, OpenAI, Anthropic, etc.) |
| Agent | The brain — holds instructions, tools, and the LLM config |
| Agent Chat | The interface — sends messages and receives responses |
For more advanced setups you can add:
- Agent Tool — Give the agent abilities (search, calculate, call APIs)
- Agent Memory — Persist conversation history across sessions
- Agent Workflow — Chain multiple steps (sequential, parallel, or conditional)
- Multi-Agent Network — Multiple agents collaborating via supervisor, round-robin, or broadcast
- MCP Server — Connect to external MCP tool servers
Available Nodes
| Node | Description | Guide |
|---|---|---|
| LLM Provider | Configure which AI model to use | LLM Provider |
| AI Agent | Create an agent with instructions and tools | Agent |
| Agent Chat | Send messages and receive responses | Agent Chat |
| Agent Tool | Define tools the agent can call | Agent Tool |
| Agent Memory | Store conversation history | Memory |
| Agent Workflow | Build step-based pipelines | Workflow |
| Multi-Agent Network | Coordinate multiple agents | Multi-Agent Network |
| MCP Server | Connect to MCP tool servers | MCP Server |
Getting Started
New to AI Agents? Start with the Getting Started guide to build your first chatbot in under 5 minutes.