IT Support Chatbot Leveraging Existing Help Portal Search API - n8n Workflow

• SaaS Companies & Product Teams: Seeking to automate level-1 customer support inquiries using existing documentation.


• Automation Engineers: Looking for advanced n8n templates demonstrating custom tooling and RAG alternatives.


• Technical Writers/Knowledge Managers: Who want to leverage their current support portal search functionality directly for AI responses.


• Anyone using n8n who needs a flexible, low-maintenance AI agent solution.

This automation provides a robust and cost-effective method for implementing Retrieval Augmented Generation (RAG) capabilities without the overhead of maintaining a separate vector database. Many organizations already have sophisticated search functionality built into their support documentation portals (often powered by tools like Algolia or Elasticsearch). This n8n workflow skips the complex indexing process entirely by connecting an n8n AI Agent directly to that existing search API via a custom n8n tool. This specific n8n workflow serves as an excellent n8n template for minimizing latency, reducing maintenance, and ensuring that the AI agent uses the most current information available in the knowledge base. This powerful n8n solution ensures factual accuracy and provides references (URLs) back to the original source.

This comprehensive n8n workflow operates in two main segments: the main Chat Agent loop and the custom Knowledgebase Tool subworkflow.

1. User Initiation (n8n Trigger): The process begins with the 'When chat message received' n8n trigger, which captures the user's initial query and initiates the chat sequence.


2. Agent Orchestration: The query is routed to the 'AcuityScheduling Support Chatbot' n8n node (a LangChain Agent). This agent uses the 'OpenAI Chat Model' (LLM) and 'Simple Memory' to determine the best course of action.


3. Tool Execution: If the agent deems that external information is required (as per its system instructions), it utilizes the custom-defined 'Knowledgebase Tool'. This tool calls a subworkflow to perform a search, passing the user's query (query) as input.


4. Subworkflow Search (HTTP Request): The subworkflow uses an 'Execute Workflow Trigger' (which is technically an n8n node acting as the entry point) and immediately executes the 'Acuity Support Search API' n8n node (an HTTP Request). This node POSTs the search query to the external support portal's Algolia API.


5. Result Check: The 'Has Results?' n8n node checks if the API returned any relevant articles.


6. Data Processing: If results are found, the 'Results to Items' n8n node splits the hits into individual items. The 'Extract Relevant Fields' n8n node cleans up the data, extracting essential fields like title, body, and constructing the full public URL, which is crucial for the final LLM response.


7. Aggregation: The 'Aggregate Response' n8n node collects all cleaned search result data into a single array named response and returns it to the AI Agent.


8. Final Answer Generation: The main 'AcuityScheduling Support Chatbot' n8n agent receives the context from the tool and generates a factual, well-referenced response using the LLM. This entire seamless n8n workflow provides a fast, effective support solution.

To install and deploy this advanced n8n workflow, follow these steps:

1. Import the n8n Workflow: Copy the provided JSON data and paste it directly into your n8n instance using the 'New' -> 'Import from JSON' function.


2. OpenAI Credentials: Locate the 'OpenAI Chat Model' n8n node. You will need to set up or select your existing OpenAI API credential. Ensure it has access to models like gpt-4o-mini or higher.


3. Customize the API: The core functionality relies on the 'Acuity Support Search API' n8n node. To adapt this n8n template for your own organization, you must update the URL, headers, and JSON body parameters to match your specific support portal's search API structure (e.g., your own Algolia index or internal API endpoint).


4. Configure the Agent: Review the 'AcuityScheduling Support Chatbot' n8n node. Customize the 'System Message' to align with your brand voice, specific restrictions, and support policies.


5. Activation: Once all credentials and API parameters are correctly set, activate the 'When chat message received' n8n trigger to start the agent. This powerful n8n workflow is now ready to receive chat messages.

• When chat message received (n8n Trigger): The starting n8n trigger for the chat session. It waits for incoming messages to initiate the automated process.


• AcuityScheduling Support Chatbot (LangChain Agent): The primary orchestration n8n node. It manages conversation state, determines when to use tools, and synthesizes the final response based on memory and external data. Configuration includes a detailed system message instructing it to use the knowledgebase tool and share URLs.


• OpenAI Chat Model (LangChain LLM): Provides the large language model functionality, configured to use gpt-4o-mini for cost-effective and fast inference.


• Simple Memory (Memory Buffer Window): Maintains conversation history, allowing the AI n8n agent to reference previous turns in the chat.


• Knowledgebase Tool (LangChain Tool Workflow): Defines the custom function that the n8n agent can call. It links to the subworkflow identified by the current n8n workflow ID, passing the user's search intent as the query input parameter.


• Acuity Support Search API (HTTP Request n8n node): The crucial n8n node that executes the external API call. It dynamically constructs a JSON POST body containing the user's query to search the AcuityScheduling Algolia index for relevant articles.


• Has Results? (If n8n node): A core logic n8n node that conditionally branches the workflow based on whether the search API returned any hits.


• Extract Relevant Fields (Set n8n node): Cleans and normalizes the API response, specifically extracting the title, body_safe, and calculating the full documentation url for optimal token usage and presentation to the LLM. This step optimizes the efficiency of the n8n workflow.


• Aggregate Response (Aggregate n8n node): Gathers all individual search result items back into a single structured response object, ready to be returned from the custom tool to the main n8n agent.