This workflow enables scalable, dynamic conversations with multiple AI agents in n8n. Each agent can be configured with a unique name, model, and personality. Users can interact with one or more agents in the same chat session by using @mentions. If no mentions are used, all agents will participate in random order.
The system dynamically handles agent setup, memory, message routing, and response formatting — all without needing multiple agent nodes or complex branching. Perfect for experimentation, brainstorming, debate-style discussions, or role-based AI simulations.
Key Features
- Dynamic Multi-Agent Setup: Easily define agents with custom names, LLM models (via OpenRouter), and system prompts.
- @Mention Support: Direct your message to specific agents, or let all agents respond automatically.
- Randomized Response Flow: When no mentions are present, agents take turns in random order.
- Conversation Memory: Shared memory ensures agents stay contextually aware across multi-turn dialogues.
- Automated Response Combining: Outputs are neatly formatted and merged for easier reading.
- Flexible Scaling: Add or remove agents with a simple JSON configuration.
- Clean Workflow Design: Centralized logic avoids duplication of nodes for each agent.
