Workflow Templates

This n8n template demonstrates how to calculate the evaluation metric "Similarity" which in this scenario, measures the consistency of the agent. The scoring approach is adapted from the open-source evaluations project RAGAS and you can see the source here https://github.com/explodinggradients/ragas/blob/main/ragas/src/ragas/metrics/_answer_similarity.py How it works This evaluation works best where questions are close-ended or about facts where the answer can have little to no deviation. For our scoring, we generate embeddings for both the AI's response and ground truth and calculate the cosine similarity between them. A high score indicates LLM consistency with expected results whereas a low score could signal model hallucination. Requirements n8n version 1.94+ Check out this Google Sheet for a sample data https://docs.google.com/spreadsheets/d/1YOnu2JJjlxd787AuYcg-wKbkjyjyZFgASYVV0jsij5Y/edit?usp=sharing

🧨 VIP Radar: Instantly Spot & Summarize High-Value Shopify Orders with AI + Slack Alerts Automatically detect when a new Shopify order exceeds $200, fetch the customer’s purchase history, generate an AI-powered summary, and alert your team in Slack—so no VIP goes unnoticed. 🛠️ Workflow Overview | Feature | Description | |------------------------|-----------------------------------------------------------------------------| | Trigger | Shopify “New Order” webhook | | Conditional Check | Filters for orders > $200 | | Data Enrichment | Pulls full order history for the customer from Shopify | | AI Summary | Uses OpenAI to summarize buying behavior | | Notification | Sends detailed alert to Slack with name, order total, and customer insights | | Fallback | Ignores low-value orders and terminates flow | 📘 What This Workflow Does This automation monitors your Shopify store and reacts to any high-value order (over $200). When triggered: It fetches all past orders of that customer, Summarizes the history using OpenAI, Sends a full alert with context to your Slack channel. No more guessing who’s worth a closer look. Your team gets instant insights, and your VIPs get the attention they deserve. 🧩 Node-by-Node Breakdown 🔔 1. Trigger: New Shopify Order Type**: Shopify Trigger Event**: orders/create Purpose**: Starts workflow on new order Pulls**: Order total, customer ID, name, etc. 🔣 2. Set: Convert Order Total to Number Ensures the total_price is treated as a number for comparison. ❓ 3. If: Is Order > $200? Condition**: $json.total_price > 200 Yes** → Continue No** → End workflow 🔗 4. HTTP: Fetch Customer Order History Uses the Shopify Admin API to retrieve all orders from this customer. Requires your Shopify access token. 🧾 5. Set: Convert Orders Array to String Formats the order data so it's prompt-friendly for OpenAI. 🧠 6. LangChain Agent: Summarize Order History Prompt**: "Summarize the customer's order history for Slack. Here is their order data: {{ $json.orders }}" Model**: GPT-4o Mini (customizable) 📨 7. Slack: Send VIP Alert Sends a rich message to a Slack channel. Includes: Customer name Order value Summary of past behavior 🧱 8. No-Op (Optional) Used to safely end workflow if the order is not high-value. 🔧 How to Customize | What | How | |--------------------------|----------------------------------------------------------------------| | Order threshold | Change 200 in the If node | | Slack channel | Update channelId in the Slack node | | AI prompt style | Edit text in LangChain Agent node | | Shopify auth token | Replace shpat_abc123xyz... with your actual private token | 🚀 Setup Instructions Open n8n editor. Go to Workflows → Import → Paste JSON. Paste this workflow JSON. Replace your Shopify token and Slack credentials. Save and activate. Place a test order in Shopify to watch it work. 💡 Real-World Use Cases 🎯 Notify sales team when a potential VIP buys 🛎️ Prep support reps with customer history 📈 Detect repeat buyers and upsell opportunities 🔗 Resources & Support 👨‍💻 Creator: Yaron Been 📺 YouTube: NoFluff with Yaron Been 🌐 Website: https://nofluff.online 📩 Contact: Yaron@nofluff.online 🏷️ Tags #shopify, #openai, #slack, #vip-customers, #automation, #n8n, #workflow, #ecommerce, #customer-insights, #ai-summaries, #gpt4o

How it works Triggers on submitting an n8n form Uses the form details to prepare a message Sends the message to Slack Set up Steps Add in your team name Add in message tone Set up Open AI Set up Slack

Find OAuth URIs with AI Llama Overview: The AI agent identifies: Authorization URI Token URI Audience Methodology: Confidence scoring is utilized to assess the trustworthiness of extracted data: Score Range: 0 < x ≤ 1 Score Granularity: 0.01 increments Model Details: Leveraging the Wayfarer Large 70b Llama 3.3 model. How it works: This template is designed to assist users in obtaining OAuth2 settings using AI-powered insights. It is ideal for developers, IT professionals, or anyone working with APIs that require OAuth2 authentication. By leveraging the AI agent, users can simplify the process of extracting and validating key details such as the authorization_url, token_url, and audience. Set up instructions: 1. Configuration Nodes Structured Output Node**: Parses the AI model's output using a predefined JSON schema. This ensures the data is structured for downstream processing. Code Node**: If the AI model’s output does not match the required format, use the Code node to re-arrange and transform the data. Example code snippets are provided below for common scenarios. 2. AI Model Prompt The prompt for the AI model includes: A detailed structure and objectives of the query. Flexibility for the model to improvise when accurate results cannot be determined. 3. Confidence Scoring The AI model assigns a confidence score (0 < x ≤ 1) to indicate the reliability of the extracted data. Scores are provided in increments of 0.01 for granularity. Adaptability Customize this template: Update the AI model prompt with details specific to your API or OAuth2 setup. Adjust the JSON schema in the Structured Output node to match the data format. Modify the Code logic to suit the application's requirements.

In this guide, we’ll walk you through setting up an AI-driven workflow that automatically processes highly-rated food photos from a Google Sheet, generates AI-powered captions, shares them to Pinterest, and updates the sheet to reflect the posts. Ready to automate your food photo sharing? Let’s dive in! What’s the Goal? Automatically detect and process highly-rated food photos (4 stars or above) from a Google Sheet. Use AI to generate engaging and relevant captions. Share the photos with captions to Pinterest via the Pinterest API. Update the Google Sheet to mark photos as posted. Enable scheduled automation for consistent posting. By the end, you’ll have a self-running system that shares your best food photos effortlessly. Why Does It Matter? Manual photo sharing is time-consuming and inconsistent. Here’s why this workflow is a game changer: Zero Human Error**: AI ensures consistent captions and posting accuracy. Time-Saving Automation**: Automatically handle photo sharing, boosting efficiency. Scheduled Posting**: Maintain a regular presence on Pinterest without manual effort. Focus on Creativity**: Free your team from repetitive posting tasks. Think of it as your tireless social media assistant that keeps your Pinterest feed vibrant. How It Works Here’s the step-by-step magic behind the automation: Step 1: Trigger the Workflow Detect new photos to post using the Daily Post Scheduler node (e.g., once daily). Initiate the workflow at a scheduled time to check for new food photos. Step 2: Fetch Food Photos from Sheet Retrieve rows from the Google Sheet that contain food photo metadata like image URLs, ratings, and status. Step 3: Filter 4+ Star Dishes Filter only those food entries with high ratings (4 stars or above) and unposted status. Step 4: AI Caption Generator Use AI (e.g., GPT/OpenAI) to create engaging and relevant captions for selected food photos. Step 5: Upload to Pinterest Automatically post the food photo with the generated caption to Pinterest via the Pinterest API. Step 6: Mark as Posted in Sheet Update the Google Sheet to reflect that the photo has been successfully shared. How to Use the Workflow? Importing a workflow in n8n is a straightforward process that allows you to use pre-built workflows to save time. Below is a step-by-step guide to importing the Automated Food Photo Sharing workflow in n8n. Steps to Import a Workflow in n8n Obtain the Workflow JSON Source the Workflow: Workflows are shared as JSON files or code snippets, e.g., from the n8n community, a colleague, or exported from another n8n instance. Format: Ensure you have the workflow in JSON format, either as a file (e.g., workflow.json) or copied text. Access the n8n Workflow Editor Log in to n8n (via n8n Cloud or self-hosted instance). Navigate to the Workflows tab in the n8n dashboard. Click Add Workflow to create a blank workflow. Import the Workflow Option 1: Import via JSON Code (Clipboard): Click the three dots (⋯) in the top-right corner to open the menu. Select Import from Clipboard. Paste the JSON code into the text box. Click Import to load the workflow. Option 2: Import via JSON File: Click the three dots (⋯) in the top-right corner. Select Import from File. Choose the .json file from your computer. Click Open to import. Setup Notes Google Sheet Columns**: Ensure your Google Sheet includes the following columns: Image URL, Rating (numeric, e.g., 1-5), Feedback (text), Pin Title, Pin Description, Destination URL, Board ID, and Status (e.g., "Pending" or "Posted"). Google Sheets Credentials**: Configure OAuth2 settings in the Fetch Food Photos node with your Google Sheet ID and credentials. AI Model**: Set up the AI Caption Generator node with OpenAI credentials (e.g., API key). Pinterest API**: Authorize the Upload to Pinterest node with Pinterest API credentials (e.g., Bearer Token) and obtain the Board ID. Scheduling**: Adjust the Daily Post Scheduler node to your preferred posting time (e.g., daily at 9 AM).

Turn any LinkedIn post into a personalized cold email opener that sounds like a human wrote it in seconds. Whether you're in sales, partnerships, or outreach, this tool reads LinkedIn posts like a human, distills the core message, and gives you a smart, conversational opener to kick off the relationship the right way. How It Works: 1.) Paste the post + author info into a short form. 2.) AI reads the post like a B2B sales expert would. 3.) Output = personalized opener, company name, prospect’s name, and next steps. 4.) Copy-paste into your cold email and hit send. The opener isn’t generic fluff — it references real details, sounds natural, and shows you actually paid attention. Perfect For: SDRs and BDRs Agency outreach Partnership prospecting Any cold outreach that starts with a real conversation Setup Steps Setup time: ~2-3 mins 1.) Add your OpenAI credentials (or use n8n’s built-in credits). 2.) Open the form and test it with the sample post. 3.) Tweak the AI prompt if you want to target a different niche or tone. (Optional) Connect to Google Sheets, a CRM, or your email tool. You're live.

This n8n template demonstrates an approach to image embeddings for purpose of building a quick image contextual search. Use-cases could for a personal photo library, product recommendations or searching through video footage. How it works A photo is imported into the workflow via Google Drive. The photo is processed by the edit image node to extract colour information. This information forms part of our semantic metadata used to identify the image. The photo is also processed by a vision-capable model which analyses the image and returns a short description with semantic keywords. Both pieces of information about the image are combined with the metadata of the image to form a document describing the image. This document is then inserted into our vector store as a text embedding which is associated with our image. From here, the user can query the vector store as they would any document and the relevant image references and/or links should be returned. Requirements Google account to download image files from Google Drive. OpenAI account for the Vision-capable AI and Embedding models. Customise this workflow Text summarisation is just one of many techniques to generate image embeddings. If the results are unsatisfactory, there are dedicated image embedding models such as Google's vertex AI multimodal embeddings.

Who is this for? Jira users who want to automate the generation of a Lessons Learned or Retrospective report after an Epic is Done. What problem is this workflow solving? / use case Lessons Learned / Retrospective reports are often omitted in Agile teams because they take time to write. With the use of n8n and AI this process can be automated. What is this workflow doing Triggers automatically upon an Epic reaching the "Done" status in Jira. Collects all related tasks and comments associated with the completed Epic. Intelligently filters the gathered data to provide the LLM with the most relevant information. Utilizes an LLM with a structured System Message to generate insightful reports. Delivers the finalized report directly to your specified Google Docs document. Setup Create a Jira API key and follow the Credentials Setup in the Jira trigger node. Create credentials for Google Docs and paste your document ID into the Node. How to customize this workflow to your needs Change the System Message in the AI Agent to fit your needs.

Use case This workflow enables a Telegram bot that can: Accept speech input in one of 55 supported languages Automatically detect the language spoken and translate the speech to another language Responds back with the translated speech output. This allows users to communicate across language barriers by simply speaking to the bot, which will handle the translation seamlessly. How does it work? Translation In the translation step the workflow converts the user's speech input to text and detects the language of the input text. If it's English, it will translate to French. If it's French, it will translate to English. To change the default translation languages, you can update the prompt in the AI node. Output In the output step, we provide the translated text output back to the user and speech output is generated in the translated language. Setup steps Obtain Telegram API Token Start a chat with the BotFather. Enter /newbot and reply with your new bot's display name and username. Copy the bot token and use it in the Telegram node credentials in n8n. Update the Settings node to customize the desired languages Activate the flow Full list of supported languages All supported languages:

This n8n template demonstrates how you can leverage existing support site search to power your Support Chatbots and agents. Building a support chatbot need not be complicated! If building and indexing vector stores or duplicating data isn't necessarily your thing, an alternative implementation of the RAG approach is to leverage existing knowledge-bases such as support portals. In this way, document management and maintenance of your support agent is significantly reduced. Disclaimer: This template example uses AcuityScheduling's help center website but is not associated, supported nor endorsed by the company. How it works A simple AI agent is connected with chat trigger to receive user queries. The AI agent is instructed to fetch information from the knowledge-base via the attached custom workflow tool (aka "knowledgebase tool"). There is no step to replicate the entire support articles database into a vector store. You may choose not too because of time, cost and maintainence involved. Instead, the tool leverages the existing support portal's search API to retrieve knowledge-base articles. Finally, the search results are formatted before sending an aggregated response back to the agent. How to use? Customise the subworkflow to work with your own support portal API and format accordingly. Try the following queries How do I connect my icloud to acuityScheduling? How do I download past invoices for my Acuity account? Requirements OpenAI for LLM. If your organisation's APIs require authorisation, you may need to add custom credentials as necessary. Customising this workflow Add additional tools to reach other parts of your internal knowledgebase. Not using OpenAI? Feel free to swap but ensure the LLM has tools/function calling support.

This workflow demonstrates a simple way to run evals on a set of test cases stored in a Google Sheet. The example we are using comes from an info extraction task dataset, where we tested 6 different LLMs on 18 different test cases. You can see our sample data in this spreadsheet here to get started. Once you have this working for our dataset, you can plug in your own test cases matching different LLMs to see how it works with your own data. How it works: It loads test cases from Google Sheets. For each row in our Google Sheet, it grabs the source document, converting it to text. Our "LLM judge" passes the input/output of each LLM to GPT-4.1 to evaluate each test case (Pass/Fail + Reason). It logs the outcome to a Google Sheet. A 0.5s pause between each request gets around OpenAI's API rate limits. Set up steps: Add your credentials for Google Sheets, Google Drive, and OpenRouter. Make a copy of the original data spreadsheet so that you can edit it yourself. You will need to plug your version in the Update Results node to see the spreadsheet update on each run of the loop.

Automate LinkedIn engagement without sounding like a bot. This workflow: 🌍 Detects language & tone (German / English) 👍 Chooses the right reaction (like / celebrate / support …) 🗣 Generates a personalised comment in your voice and mentions the author 📲 Optional Telegram review – approve ✅ or regenerate ❌ before posting 💸 Runs on cost-efficient GPT-4o mini or Claude 3.5 Haiku ☁️ Publishes comment + reaction via the Unipile API Setup (≈ 15-30 min) Unipile – connect LinkedIn → copy account_id, dsn, then create an Access-Token (X-API-KEY). Telegram (optional) – create a bot, add a credential named YOUR TELEGRAM ACCOUNT. OpenAI / Anthropic – add your API key and keep one LLM node (delete the other). Open the “Defining guardrails” node and replace the credential placeholders. (Optional) Tweak role, comment_length and openers_example_1-3 for your brand voice. Security: no live keys included – all secrets are placeholders. Best for: solopreneurs, marketing teams, personal-branding consultants.