How to Create and Configure BoldDesk AI Agent
The BoldDesk AI Agent is an autonomous virtual assistant handling customer interactions quickly and accurately. It supports multiple AI agents tailored to different needs, making it scalable. Agents can send automatic replies triggered by user actions to improve engagement and reduce wait times. This guide covers configuring and deploying AI Agents in BoldDesk, including creating, training, customizing, channel mapping, managing multiple agents, testing, and setting permissions.
The AI Agent functionality is offered exclusively as an add-on for the Momentum and Enterprise plans.
Why Your Business Needs an AI Agent
Customers want fast, accurate help anytime. Support teams face more queries on many channels. An AI Agent gives instant, consistent 24/7 replies and handles repetitive requests, helping businesses scale support without losing quality.
Key Benefits of Deploying the BoldDesk AI Agent
- Faster Response Times
- Higher Efficiency and Resolution Rates
- Improved Customer Experience Across Channels
- Omnichannel Coverage
How to Create an AI Agent
To create an AI Agent, follow the steps below;
-
Go to AI Center > AI Agents.
-
Click on Create AI Agent.
-
In the dialog box, provide the following information.
Field Description Agent Name Here you provide the name for your AI Agent. This is only used for identity in the agent portal; however, you can use the same Agent Name as Display Name. Display Name The name shown publicly in the communication interface. Avatar An avatar is a visual representation of a user in a digital system. You can either choose from the provided avatars or upload a photo to suit your business needs using the upload icon. You can upload .jpg, .jpeg, or .png formats. Description This is optional text to use in describing your AI Agent. Brand Here, you have the option to select the brand to which the AI Agent belongs. -
Click on create to finalize the creating process.
Configuring AI Agent
Once you have created the AI Agent, you can configure the following settings;
Persona
This defines your AI Agent’s persona and brand identity with a name, avatar, business details, and instructions. You can edit the details from this page. You can also customize the following settings here;
| Setting | Description |
|---|---|
| Response Language | You can choose whether your AI Agent will respond in the customer query language or a fixed language. If you choose fixed language, you also have to choose the preferred language from supported languages. |
| Tone of Voice | Defines the communication style your AI Agent should use in conversations. You can choose from Professional, Casual, or Friendly. |
| Response Length | Choose your preferred response length for the AI Agent. |
Once you have provided all the details, click on update.
Guidance
Define how your AI Agent should act, what tasks it’s responsible for, and when it should escalate issues. This ensures consistent behaviour, decisions, and communication across all interactions. You can customise the following under guidance.
| AI Agent Guidance | Description |
|---|---|
| Agent Role | This section contains a text box where you define the overall identity, purpose, and behavioural expectations of the AI agent. |
| Agent Instructions | This section provides space to write additional rules, constraints, or operational instructions for the agent. |
| Immediate Handover | This text area is for defining when the AI should instantly hand off a conversation to a human agent. |
| Conditional Handover | This section lets you specify situations where the AI should transfer the conversation to a human based on conditions or thresholds. |
| Out of Scope | You can enable to ensure that the AI stays within the defined limits and does not respond to queries outside the allowed scope. |
| PII Governance | If enabled, it enforces rules to protect personal and sensitive information in customer queries. |
| Content Safety | Enables content safety checks to filter harmful or inappropriate material according to your preferences. You may select options for: Verbal abuse/harassment, Hate speech, or Dangerous/illegal requests. |
Once you are done click on update.
If you enable either Out of Scope, PII Governance, or Content Safety, you are required to provide a custom message for immediate handover or decline.
Knowledge Sources
Boost your AI Agent’s accuracy by providing structured knowledge sources, including articles, Q&As, websites, and files.
| Knowledge Source | Description |
|---|---|
| KB Articles | Add articles from your knowledge base to your AI Agent for comprehensive knowledge integration. |
| Q&A | Bring Q&A pairs from your collection to make your AI Agent smarter. |
| Web Pages | Add web pages from your collection so your AI Agent can provide more context. |
| Files | Include files from your collection to give your AI Agent extra resources. |
When you click on add for each of the Knowledge Sources, you can choose whether to use all content or specific. You can also manage the sources using the manage button. Explore How to Add AI Data Sources in BoldDesk.
AI Tools
AI Tools let AI Agents and Copilot execute functions, look up data, provide insights, or integrate with external systems—via APIs or Model Context Protocol (MCP). You can refer to each of the articles for each AI tool as in the table below.
| AI Tool | Reference |
|---|---|
| API AI Tool | Adding API Actions let AI Agents execute functions, look up data, provide insights, or integrate with external systems—via APIs. How to Configure API AI Tools in BoldDesk. |
| MCP Server | Linking MCP Servers let AI Agents execute functions, look up data, provide insights, or integrate with external systems—via Model Context Protocol (MCP). Explore How to Configure MCP Server AI Tools in BoldDesk. |
It is important to test your AI Agent to make sure everything’s running smoothly before going live. You can use the preview pane to test the AI Agent. Explore Customizing and Testing AI Agent Capabilities in BoldDesk.
Publishing and Deploying the AI Agent
Publishing and deploying constitute the final phase in making your AI Agent operational. After creating and properly configuring your AI Agent, it is advisable to test it before publishing. Publishing initiates the deployment process to the desired channel for your AI Agent. Upon clicking publish, a pop-up window will display the available channels for deployment. You may select one or multiple channels, and for each channel, click connect to specify the deployment location. Additionally, you can remove the AI Agent from this page if it has been previously connected to any channels.
Once you are done, click on publish and your AI Agent will be live in all the selected channels.
Where You Can Deploy the AI Agent
BoldDesk allows you to deploy your AI Agent across multiple communication touchpoints, ensuring consistent support:
- Live chat
- Web widgets
- WhatsApp (Meta Cloud API)
- Twilio SMS
- Telegram
- WhatsApp (Gupshup)
- Line
- WhatsApp (Twilio)
AI Agent features require active AI Credits to operate and will not function without AI Credits. Please ensure that credits are purchased and active to use all AI functions in BoldDesk. For more information, refer to Understanding AI Credits in BoldDesk AI Agent.
Unanswered Questions
This section is available for only published AI Agents. All questions that the particular AI Agent was unable to respond to will be listed here. Review unresolved queries from AI Agent to find content gaps, improve knowledge quality, and increase response accuracy. Explore Managing Unanswered Questions in BoldDesk.
Performance
Similar to Unanswered Questions, this section is accessible only for published AI Agents.
From this page, you can view an overview of the following metrics for the past 7 days, which can be updated using the refresh icon.
| Metric | Description |
|---|---|
| Deflection Rate | Displays number of conversations in either resolved or abandoned category by the AI Agent during the selected time period, along with the comparison to the previous period. |
| Escalation Rate | Displays the number of conversations in the escalation category by the AI Agent during the selected time period, along with the comparison to the previous period. |
| Daily Credits Consumption | Displays how many AI credits your workspace is using per day. |
| Total Credits Used | Refers to the total number of AI credits your workspace has consumed within the current billing cycle. |
From this page you can also click on built-in report links to view either AI Credits Usage Dashboard or AI Agent performance dashboard.
AI Agent Versioning
This refers to the way of managing different states or versions of an AI Agent as you create, edit, test, and deploy it. Unanswered questions and performance metrics are updated only in the published version. Any edits in a draft version won’t change these metrics.
Permission Required to Manage AI Agents
To manage AI Agents, users must have Manage AI Agents permission.
Go to Admin > AI > Roles and Permissions > Manage AI Agents
You can also configure the AI chatbot in BoldDesk; learn how to Get Started with BoldDesk AI Chatbots.
Frequently Asked Questions
Q1: Can I change the agent’s brand after creation?
A: No. Create a new agent for a different brand to avoid cross‑brand content leakage.
Q2: Why aren’t Unanswered Questions updating?
A: Only published agents collect these insights. Publish your changes first.
Q3: What’s the difference between Publish and Deploy?
A: Publish finalizes the agent version. Deploy maps that version to channels (e.g., web chat, WhatsApp).
Q4: Do I need AI Credits for testing?
A: No, AI interactions only consume credits after going live.
Q5: How do I prevent the bot from giving legal or medical advice?
A: Use Out of scope with a clear decline message and enforce Immediate handover for regulated topics.
