AI agent

Speed up your support with AI

8 articles

Enable AI agent without blowing up your support inbox

The AI agent should be enabled gradually and with careful monitoring. What we recommend is to start with a 1-3 well-tested intents, and expand only once you’ve verified that everything works as expected. Avoid letting the AI handle everything in your inbox from the start. 1. Restrict responses to enabled intents only Use the "Restrict to enabled intents" setting to control which requests the AI agent can respond to: - When checked: The agent will only respond to requests that match the enabled intents - When unchecked: The agent may attempt to answer any incoming message, even if no instructions exist. This can lead to unexpected or incorrect replies. Check that you've this checkbox selected and max. 1-3 well-tested intents enabled when starting out, and only add more once you those work as expected. You can enable and disable intents from the Intent rules section: 2. Only enable intents that have been tested in the Playground Before enabling the AI agent and any intents, test it thoroughly in the Playground using real customer conversations. Use the Replay feature to see how the AI responds to specific tickets. This allows you to confirm that it behaves as expected before going live. The AI model cannot guess what you want it to say—it will only follow the instructions you provide. Start small with simple intents. Monitor performance for a few days before expanding to more complex use cases. 3. Use a clear AI agent name and signature We recommend configuring the agent’s name and signature to make it obvious that responses are coming from an AI, not a human. This helps set expectations for your customers and reduces confusion if the AI responds incorrectly. It also encourages customers to ask for human help when needed. 4. Disable business hours and auto-reply automations Before enabling the AI agent, make sure to turn off any business hours settings and “Send message” automations (such as auto-replies). These features can trigger unnecessary messages and confuse customers when used alongside the AI agent. Good to know - Complex multi-step instructions are not yet fully supported. Improvements are coming soon. - Shopify actions, such as changing customer addresses or cancelling an order, don't work reliably yet. Improvements coming in the same update!

Last updated on Jul 25, 2025

Introduction to the AI agent

What is the AI Agent? The AI Agent is an autonomous support bot that handles customer tickets without human intervention. Built specifically for Shopify stores, it manages common support tasks that plague e-commerce stores. You can customize its tone, leverage your store data, and provide specific instructions for handling tickets. When enabled, the AI Agent is automatically assigned to new tickets and responds immediately unless configured otherwise. Human agents can remove the AI Agent from any conversation as needed. Key Features Playground Test the AI Agent safely by simulating conversations using your actual Shopify customer data. Shopify Actions The agent performs essential e-commerce tasks: - Order tracking and shipping updates - Address modifications - Order cancellations - Item replacements (size, color changes) - Order tagging - Standard helpdesk functions (ticket assignment, private notes) Knowledge & Guidance Train the agent using your store's products, store pages, help articles, canned responses, saved training responses and custom guidances for specific support scenarios. Multi-Platform Support Deploy across email, live chat, direct messages, and social media comments. Common Questions How do I prevent hallucinations? The agent is programmed to handover tickets when uncertain about available knowledge or guidance. The most effective prevention method is rigorous testing in the Playground using your historical tickets. Users who thoroughly test against old tickets see the best results. Can I limit which tickets the agent handles? Yes. Configure the AI Agent to respond only to tickets with specific guidance in the settings. This approach helps you gradually adopt the technology while maintaining control over ticket handling. What model powers the agent? We use multiple providers with proprietary technology for testing, evaluation, and tool execution. Our agent reliably follows complex, multi-step processes without the $1-per-ticket costs charged by other platforms. How do I add custom documents to the knowledge base? Copy your documentation into help center articles and leave them in draft status. Control which articles the AI Agent accesses through Knowledge settings. Alternatively, use AI training responses in Knowledge settings to add question-and-answer pairs directly. PDF and spreadsheet uploads are not currently supported. Does the agent work with multiple stores? The AI Agent functions across all stores connected to your account, but shares knowledge between them. For separate brands, use distinct Commslayer accounts (easily switchable via the top-left dropdown menu). Can the agent process customer images? Image processing is not currently available but is planned for future release.

Last updated on Aug 06, 2025

Training your AI agent

Your AI agent learns from different types of data to effectively answer customer queries. You can see all the data AI has access to by navigating to AI agent -> Knowledge sources (requires admin rights). The AI agent uses these sources to dynamically understand and respond to customer questions. AI training responses Responses you can save from good human-agent interactions. These provide the AI agent with examples of how to handle various customer questions, improving its ability to respond effectively. How to add an AI training response You can find an "Add AI training response" button by hovering over any message sent by a human agent. It opens a popup where you can edit the response (it is automatically polished by AI) and save it to your knowledge library. In Playground, you will see a "Revise answer" button when you hover over any message sent by the AI agent. You can give AI the expected correct response and save it to the training responses: Other knowledge types AI has access to: - Canned responses: your saved pre-written response templates your support team regularly uses. - Help articles: Information directly from your help center. - Shopify data: Information from your Shopify store’s pages and products. Guidance Guidance gives your AI agent specific, pre-defined responses and actions when it recognizes certain customer requests or scenarios: - For example, if a customer asks about a refund, your AI agent can automatically provide your pre-written refund instructions. - You can also define specific actions like assigning conversations to particular agents, adding labels, or automatically resolving issues. - Regularly updating Guidance helps your AI agent handle common requests efficiently, saving your team valuable time. How to edit Guidance: You can also add actions to intent rules, similarly to automations, and instruct AI when certain actions should be executed: Debugging AI responses Hover over any AI-generated response and click </> View prompt & knowledge used to see all the information and context the AI agent considered. Managing your AI training data You can manually enable or disable specific data sources. Disable outdated or irrelevant content to ensure your AI agent provides accurate responses. Best practices: - Regularly audit canned responses, help articles, and response examples. - Spend time on Playground testing how the agent responds to different queries - Add AI training responses to give additional context and improve your AI agent's replies. Following these guidelines ensures your AI agent delivers accurate, helpful, and consistent customer support.

Last updated on Jul 25, 2025

Using the Playground

What is the Playground? Playground lets you safely test your AI bot before using it with customers. You can chat freely, experiment with responses, and ensure everything works as expected—without impacting real customers. How to test your bot: Chat directly: Type questions and see how your bot responds. Replay past conversations: Import previous customer interactions using the Replay button, found below any customer message on any conversation. This helps you test how your bot would handle real-world scenarios. Understanding AI responses with the "Sources" modal: Every time the AI agent responds, you can review why and how it generated the reply. Click on the Sources icon (</>) located in the toolbar beneath every AI-generated message. A modal window will appear, showing all the knowledge and context your AI used to reply. Revising answers Revising an answer lets you add an AI training response to the knowledge library when the agent gives you misinformation or a bad response in the playground. Managing AI knowledge: If you notice issues, such as repetitive canned responses, visit the Knowledge page. There, you can easily exclude or edit problematic knowledge sources, helping your AI respond more naturally and effectively.

Last updated on Jun 09, 2025

Guidance vs. AI training responses

Understanding the difference between Guidance and AI training responses helps you maximize your AI agent's performance. Guidance - Clearly defined instructions that your AI must always follow when specific intents are detected. - Suitable for enforcing mandatory responses and actions (e.g., assigning an agent, tagging, resolving conversations). - Provide structure and consistency across frequent scenarios. Example: When customers ask for a refund, always include a link to the return page and mention the 30-day return window. Add label "refund" to the conversation. "<add_label:refund>" AI training responses - Training responses are provided to teach your AI how to reply, but the AI decides dynamically when and how to use them. - Help your AI generate accurate, natural-sounding answers. - Inserted dynamically, meaning the AI chooses the most relevant example based on the specific context. Example: Customer: "How long does shipping take?" Sample response: "Our standard shipping typically takes 3-5 business days." Quick tips: - Regularly review and update Guidance to keep AI performance sharp and aligned with your policies. - Regularly add AI training responses to enhance AI adaptability and accuracy. Keeping these clear helps ensure your AI consistently provides effective customer support.

Last updated on Jul 25, 2025

How to enable AI social media comment moderation

AI will automatically hide toxic or inappropriate comments like spam, profanity, negativity, complaints, and ads. You can review and unhide comments if needed. Steps to Enable: 1. Go to AI agent -> Social media moderation. 2. Toggle on "Auto-hide inappropriate comments". To review hidden comments, check the Hidden comments view in the sidebar.

Last updated on Jun 09, 2025

Use the AI agent with non-English languages

Our AI can reply in many languages, but it still processes instructions in English. Until we launch full multilingual embeddings, follow these guidelines: - Write all agent instructions in English. The AI translates its replies for the customer. - Keep your knowledge content in English whenever possible. A single source language helps the agent respond faster and more accurately. Multilingual embeddings are coming soon. When they’re available, you’ll be able to give instructions and knowledge in any supported language.

Last updated on Jul 10, 2025

How to setup the AI agent

Enable Shopify permissions If you installed Commslayer before August 2025, you must reauthorize your Shopify connection to grant order editing permissions. 1. Navigate to Settings → Shops 2. Click "Reauth" if displayed 3. Complete Shopify authentication and confirm updated permissions Customize agent settings Configure the AI agent's identity and behavior through basic customization options. Agent name Most brands include "AI agent" in the name (e.g., "Awesome Brand's AI agent"). Some merchants, particularly dropshippers, use human names without AI disclosure. Your choice. General rules (optional) Define behavioral guidelines for the AI agent. Only add rules after thorough playground testing and guidance configuration. Tone of voice (optional) Specify communication style per channel (friendly, formal, brief, Gen Z, emoji-free). The agent defaults to channel-appropriate helpful tones. Change only if dissatisfied with default behavior. Signature Set email and direct message signatures. Useful for AI disclosure to customers. Handover settings Configure whether the AI agent transfers uncertain conversations to unassigned queue or specific human agents. Set the fallback message sent during handovers. Enable the AI agent Access AI agent settings from the left menu to select active inboxes. Restrict to enabled intents Critical setting that appears after inbox selection: - Disabled: AI agent handles all inbox tickets using available knowledge and guidance - Enabled: AI agent only responds to tickets with matching guidance, regardless of other available knowledge This restriction helps gradual adoption by limiting responses to specific scenarios like order tracking or cancellation requests. Operational details - Response delays: The AI agent includes randomized response delays for email and social media comments to appear natural and prevent customer frustration with instant replies. - Assignment behavior: The AI agent is automatically assigned to new tickets. Removing it stops responses to that conversation. If initially disabled for a conversation, the agent won't respond to existing messages until new customer messages arrive.

Last updated on Aug 07, 2025