AI Chat which response based on employee profile

[weilies_chok]

i have a working AI chatbot as below and it’s working. It can pull from all different agent (IT, JIRA, Interface, event pull/schedule event)

How we can further extend it if i need the bot only show under our ERP, after emp logged in? I understand how to share out the html widget to show the chatbot after login screen.

• How can i tell the chatbot, you are currently interact with “EMP01”, not other employee? I need the interaction to be specific to EMP01
• How to push in “Role”, “EmployeeCode” into the chatbot, if detected the Role=IT, the chatbot allows the “IT Agent” to respond. otherwise return “no result”. Basically a design that control which agent the user can interact with based on role
• I will extend another “Create Leave Request” API, but still the API should create leave for that specific employee. How to extend that?

i tried to search youtube vid but can’t seems to find any constructive resource.

Information on your n8n setup

• n8n version: 1.67.1
• Database (default: SQLite): qdrant
• n8n EXECUTIONS_PROCESS setting (default: own, main): default
• Running n8n via (Docker, npm, n8n cloud, desktop app): docker
• Operating system: window

[ria]

Hi @weilies_chok

You can use the Agent’s system message to populate “static context” like this. In n8n this can be populated dynamically from the input using expressions, see screenshot below for an example of what I mean. You can then flesh out your instructions further based on what the value is.

Check out this tutorial we’ve recently released

Building AI Agents: Chat Trigger, Memory, and System/User Messages Explained [Part 1]

Hope it helps!

[mredodos]

if you not need login or something like that you can put some data from your source or from a database for example and put inside the prompt with expressions,and with prompt you can force the agent to answer or not