Build Smarter Support in Minutes
Multi agent systems used to be something only enterprise platforms could handle. ChatLogic makes it simple for any WooCommerce store. By combining dedicated agents, intent routing and targeted knowledge files you can automate a large part of your support workflow without sacrificing quality. This guide walks you through setting up your first multi agent system from scratch.
What a Multi Agent System Does
Instead of one chatbot trying to answer every question, ChatLogic splits responsibilities across specialized agents. Each agent has its own instructions, tone, knowledge and model choice. When a customer sends a message, ChatLogic identifies the intent and routes it to the agent best suited to respond. This leads to accurate answers, faster response times and more reliable automation.
Examples of Agent Types
- Orders and delivery
- Returns and refunds
- Product information and recommendations
- Billing or payment questions
- General store support
Step 1 - Install and Connect OpenAI
Start by installing the ChatLogic plugin inside your WordPress dashboard. Once activated, enter your model API key and choose a default model. GPT 4o is recommended for most stores thanks to its balanced speed, cost and accuracy. You can always assign different models to different agent types later.
Step 2 - Create Specialized Agents
Navigate to the Agents section to create your first set of roles. You can use the built in templates for Orders, Returns or Product Info or build a custom agent from scratch. Each agent has its own prompt, behaviour style and access to files. This separation keeps answers consistent and prevents cross topic confusion.
What You Can Configure Per Agent
- Agent name and personality
- Model choice
- Instruction prompt
- Knowledge files
- Tone and behaviour settings
Step 3 - Upload Knowledge Files
Each agent can have its own knowledge base. Upload PDFs, TXT documents or JSON files that contain product manuals, shipping tables, store policies or FAQ content. ChatLogic automatically summarises and indexes these so the agent can retrieve answers quickly. This keeps your content accurate even as your store evolves.
Tips for Effective Knowledge Management
- Use simple files rather than long complex PDFs.
- Separate content by topic to keep agents focused.
- Update files whenever store policies change.
- Use product JSON files to serve fast retrieval for catalog data.
Step 4 - Test and Refine
Open the built in tester and send sample questions to see how ChatLogic routes messages. You will notice that each agent responds only to queries that belong to its topic. If routing seems off you can adjust the agent prompts or add intent keywords to improve accuracy.
Examples of Test Prompts
- "Where is my order?"
- "I want to return this item."
- "Does this product come in size XL?"
- "What payment methods do you support?"
Step 5 - Enable Intent Routing
Intent routing ensures the right agent handles each message without manual tagging. ChatLogic analyses wording, context and sentiment to understand the user’s intent. Once activated your system becomes fully automatic.
Pro Tip
Turn on sentiment detection to escalate frustrated users directly to your human support channel. This keeps conversations calm and prevents negative reviews.
Going Live
Once you are satisfied with the agent setup, enable the chat widget on your WooCommerce store. The system will now autonomously route questions, answer instantly and use your uploaded knowledge files to stay accurate.