How to create a Personal Agent in n8n

In the AI ​​era of 2025–2026, there are dozens of tools that allow you to create chatbots or AI assistants. But what sets Personal Agent in n8n apart is that your agent doesn't just chat – it can execute actions through over 400 built-in integrations, from sending emails and updating CRMs to querying databases and calling any API.

The Personal Agent in n8n is accessed directly through the Chat interface in the dashboard – you don't need to open the workflow or click "Execute," just send a message like you would with ChatGPT . This is a relatively new feature, appearing alongside n8n's direction of transforming the platform into a central AI agent for the entire team, not just an automation tool.

Step-by-step guide: Creating a Personal Agent in n8n

Step 1: Open the Chat interface

On the main n8n dashboard, find and click the "Chat" tab in the left-hand navigation panel. This is where you manage all your chats and agents.

Step 2: Select the Personal Agents tab

In the Chat interface, you will see two main tabs. Select the "Personal Agents" tab to view a list of created agents and create a new agent by selecting New Agent.

Steps 3+4: Configure the icon and agent name.

Click on the "Icon" menu to select an icon to represent the agent - this is a feature that helps you quickly distinguish between agents when there are many agents in the list.

Next, name the agent. The name should clearly describe its function, for example: "Email Marketing Assistant", "Data Analytics Agent", or "Research Assistant".

Enter a brief description of the agent in the description box. This helps you and your colleagues quickly understand what the agent does without having to read the entire prompt.

Steps 5+6: Write System Prompt

This is the most important step that determines the quality of the agent. The system prompt describes:

1. The role and expertise of an agent
2. How should the agent respond (tone, format, language)?
3. What agents should and shouldn't do.
4. Organizational context or use case

Here's an example of an effective prompt:

Bạn là trợ lý marketing nội bộ của [tên công ty]. Nhiệm vụ của bạn là giúp team tạo nội dung, phân tích hiệu suất campaign, và đề xuất ý tưởng dựa trên dữ liệu thực tế. Luôn trả lời bằng tiếng Việt. Khi được hỏi về số liệu, hãy truy vấn database trước khi trả lời thay vì đoán.

Step 7: Select Model

Click on the "Model" menu to select the LLM that will run the agent. n8n supports multiple providers:

1. OpenAI (GPT-4o, GPT-4o mini)
2. Anthropic Claude (Claude 3.5 Sonnet, Claude 3 Haiku)
3. Google Gemini
4. Local model via Ollama

Choose a model that strikes a balance between quality and cost, suitable for your use case.

Step 8: Add Tools

Click on the "Tools" menu to add tools the agent can use. This is the strongest point of n8n Personal Agent - you can connect the agent to other workflows, specific nodes, or entire n8n integrations. For example, add the "Supabase" tool so the agent can query the database, or the "Gmail" tool so the agent can send emails when needed.

Step 9: Save and use

Click "Save" to create the agent. The agent will appear in the Personal Agents tab and be ready to chat immediately.

Personal Agent vs. Traditional AI Workflow: When to Use Which?

The Personal Agent in n8n is designed for conversational interaction – you ask, the agent answers, and acts according to the context of the conversation. This differs from a typical AI Workflow (using Chat Triggers + AI Agent nodes) which is event-triggered and performs a fixed sequence of steps.

Use a Personal Agent when: you need an assistant who can handle many different types of requests within the same conversation, or you need context to be maintained throughout multiple Q&A sessions.

Use AI Workflow when: you need a fixed pipeline that runs automatically when an event occurs, and you need precise control over each processing step and branching condition.

Advantages and disadvantages

Advantage

1. Simple interface, no need to open workflows: End users only need to chat, without needing to understand workflow automation or node configuration.
2. Powerful tool integration: Agents can call any n8n integration as a tool – this is a major advantage over pure chatbots that cannot perform actions.
3. Manage multiple agents for multiple purposes: You can create separate agents for each team, each use case, with different system prompts and toolsets.

Disadvantages

1. It's also a relatively new feature: the Personal Agent interface in n8n is considered a feature under development - some advanced agent management features (e.g., granular permissions, agent-based analytics) are not yet fully complete.
2. LLM cost depends on the number of users in the selected model: Each chat session consumes tokens of the chosen model. For teams with many frequent agent users, the LLM cost needs to be carefully calculated.
3. System prompts require time investment: The quality of the agent directly depends on the quality of the system prompt. Writing a good prompt for a specialized agent requires extensive testing and refinement.

Conclude

Personal Agent in n8n is a significant step forward in transforming automation platforms into truly actionable AI assistants. With the ability to connect to over 400 applications as tools, it's not just a chatbot – it's an agent that can perform real tasks. If your team is already using n8n for automation and wants to add a conversational AI layer without adding new tools, Personal Agent is the ideal starting point.