# Help Desk Support / ITSM Automation via External Chat Browser
### 1.Use Case Overview
#### Objective:
Enable IT support teams to automatically respond to IT-related issues (e.g., “Printer not working”) using the Soika AI Agent. The agent can collect user credentials, search knowledge bases, and escalate by raising a support ticket and assigning a technician based on the severity — all automated using natural language instructions.
#### Primary Actors:
* End User (Employee/Staff reporting an issue)
* Knowledge Base (PDFs, Confluence pages, internal documentation)
* Base row Database (Technician Directory + Ticket Management)
#### Target Platforms:
* WhatsApp
* Email
* External Web Chat Dashboard
* Internal IT Knowledge Base
* Base row (Database for Ticket & Technician info)
#### Business Value Proposition
The Soika AI HelpDesk agent helps organizations reduce IT support load by resolving common issues like “Printer not working,” “VPN not connecting,” or “Outlook not syncing” using predefined solutions. It minimizes response time, triages effectively, and ensures escalations are handled swiftly by qualified technicians.
#### Benefits:
* 80% faster query resolution
* Knowledge base leveraged automatically
* Auto-ticket creation with technician assignment
* Reduced support costs and manual triage
* Support via natural language (Email, WhatsApp)
### 2. Flow Summary
User gives prompt: *"* Create a General Agent which is a Customer Support Assistant integrated. Your role is to receive incoming queries by internal chat and manage them systematically using Base Row DB and Emails.”
* Soika creates an AI Agent tailored to Heldesk.
* AI Agent connects to Email Tool and Base Row DB tools via Authorization from Default tools.
* Tools request OAuth authorization (user grants access).
· User Prompt to the Agent: “Hi, my printer is not working.”
System Response:\
→ AI Agent: “Please confirm your email and password for verification.”\
→ User provides credentials.\
→ Agent searches internal KB for known solutions.\
→ If not resolved: Agent checks Base row DB to find an available technician.\
→ AI Agent raises a support ticket (with severity = Medium or High).\
→ AI informs user: “Your case has been assigned to our technician \[Name]. You will be contacted in the next 30 minutes.”
### 3. Step-by-step process
### A. AI Agent Creation
SL.NO
STEPS TO BE FOLLOWED
Step 1
User opens Soika Mockingjay platform with Login ID and Password
Step 2
Ask Soika to Create an AI Agent based on the use case with Natural Language Prompt
Example : Create a General Agent which is a Help Desk Support Assistant integrated. Your role is to receive incoming queries by internal chat and manage them systematically using Base Row DB and Emails.
Step 3
The Agent is created in the left panel : Help Desk Support Assistance
Choose your LLM Model from the Agent in Edit AI-> AI Model-> llama3.3
Step 4
Once Agent is created, you can upload your knowledge Base, by Clicking on
Edit AI--> Instructions -> Knowledge Base ->Add Knowledge Base
Step 5
Once added, Enable it with the help of toggle icon. Upload any Knowledge base based on your organization needs (E.g. IT_Support_Levels_with_Queries_and_Solutions.xlsx)
Soika created the Help Desk Support Assistance Agent
Step 8 :Enable the functionalities of Base row DB and Email
### Accessing the External Chat Browser
Step I : Click on Edit AI-> General -> Change Visibility to Public -> Preview
Click on Preview
Step II : Interact with the Agent via Speech or Text
Use the microphone Icon for Voice communication
Step III : Agent accessing the Knowledge base for your query
Step IV : Automatic ticket Generation Example : Ticket- 123456, if your query is not resolved.
Tools used : Base Row DB to get the Customer support executive details.
Tools used by Agent for Automatic Ticket Creation
### **Example of system instructions you can provide**
You are an AI-Powered ITSM Support Assistant integrated into an internal chat system. Your role is to manage incoming queries systematically using Base Row DB and Emails. Follow the steps below to handle support queries efficiently. Ensure that the output does not contain any XML tags.
**Step 1: Receive and Interpret Query**
\- Upon receiving a message such as “I have an issue with the system and need help ASAP,” analyze the text to extract the core issue.
\- Ask the customer for their email ID and phone number.
\- Check the IT Support Knowledge Base (IT\_Support\_Levels\_with\_Queries\_and\_Solutions.xlsx) for known issues and resolutions.
\- Provide an initial resolution based on the knowledge base. For example, suggest actions like checking router status, restarting the PC, verifying credentials, or resetting the adapter.
**Step 2: Redirect to Base Row DB**
\- If the user is unsatisfied with the solution from the knowledge base, redirect them to the BASE ROW DB Table with Table ID (609727).
\- Use the table fields: Severity Level, Contact Person, and Contact Details.
**Step 3: Escalation Handling**
\- If the user indicates that the solution did not work, generate a random ticket number as only numbers.
\- Add the generated ticket number in the column of TICKET\_NO, QUERIES and SEVERITY in a new row in the Base Row DB Table ID (610737).
\- Based on the severity and queries in the Base Row DB Table (610737), assign the case to L1/L2 Support as detailed in Base Row Table ID (609727).
\- Set the priority to High, due within 30 minutes, and label it as Needs Technician Visit.
**Step 4: Retrieve Technician Contact Info**
\- Access Base Row DB (Table ID: 609727) using the escalation tag (e.g., L1 or L2).
\- Also, access the Base Row DB (Table ID: 610737) to fetch the ticket number.
\- Extract the contact details of the assigned technician, including Technician Name, Contact Email, Phone, and Specialty.
\- Send an email to the respective technician.
**Step 5: Notify Customer**
\- Send an email to the customer with the ticket number, stating: “We’ve assigned your case to our senior technician, \[Contact Name]. You can reach them at \[Email]. They’ll begin resolving your issue within the next 30 minutes.”
**BASE ROW DB Table with Table ID (610737)**
Tool
Purpose
Soika Agent
Natural language assistant
Base row
Ticket DB + Technician Directory
Internal Docs
Knowledge base for common issues
Email/WhatsApp
Input channel for users
Slack/Lark API
Admin notifications (optional)
### User and Agent Interaction
Email Output received by USER with Ticket Number
---
# Agent Instructions: Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://soika-labs.gitbook.io/soika-mockingjay/use-cases/help-desk-support-itsm-automation-via-external-chat-browser.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.