Skip to content
Gary edited this page Jan 19, 2026 · 3 revisions

๐Ÿš€ Welcome to the NeroiStack Wiki!

Welcome to NeroiStack, a powerful agentic AI framework. To get started with your first AI chat instance, please follow the steps below.


๐Ÿ›  Getting Started Guide

1. Configure LLM API Keys

Before chatting, you need to provide an API key for your preferred AI provider (e.g., OpenAI, Google Gemini, Azure OpenAI, or Local Ollama).

  • Go to the Key Manage page.
  • (Optional) Select your provider and paste your API key.
  • (Optional) Configure the Custom Endpoint if you are using a proxy or local service.
Key Management Screenshot

2. Define Your Agents

Agents are the "personalities" or "tools" of your system. You can define multiple agents with different instructions.

  • Navigate to the Agents page.
  • Click Add Agent.
  • Set the System Instructions (e.g., "You are a professional Coding Master").
  • Configure model parameters like Temperature and Max Tokens.
image

3. Create a Chat Configuration (Chat Class)

A Chat Configuration defines how agents work together (Orchestration).

  • Go to the Chat Manage (Chat Class) page.
  • Click Add Chat Config.
  • Select an Orchestration Strategy:
    • Single: One agent handles everything.
    • Sequential/Group Chat: Multiple agents collaborate.
  • Link the Agents you created in Step 2 to this configuration.
image

4. Launch a Chat Instance

Now you are ready to talk!

  • In the main chat interface or the Chat Hub, click Add Chat Instance.
  • Select the Chat Class you defined in Step 3.
  • Give your session a name and start chatting! ๐Ÿ’ฌ

๐Ÿ’ก Pro Tips

  • Model Switching: You can change the AI model on the fly within a chat instance using the model dropdown.
  • MCP Integration: If you have MCP (Model Context Protocol) servers, you can add them as plugins to your agents to give them real-world capabilities.

Tip

Need help? Feel free to open an issue or check the Discussions page!