Enrich Property Inventory Survey with Image Recognition and AI Agent n8n Workflow Template

n8n workflow template · 19 nodes · OpenAI

Enrich Property Inventory Survey with Image Recognition and AI Agent is a free n8n workflow template that runs when another workflow calls it and works with OpenAI and Airtable. It has 19 nodes, 6 of them AI nodes (OpenAI Chat Model, Call n8n Workflow Tool, Structured Output Parser and OpenAI). The notes inside the workflow say: “Join the Discord or ask in the Forum!” To run it, you add your own credentials for Airtable API, HTTP Header Auth, OpenAI API and Serp API.

Download JSON How to import into n8n

At a glance

Trigger
Execute Workflow Trigger (Execute Workflow Trigger)
Nodes
19
Apps
OpenAI and Airtable
AI nodes
6
Credentials
4 types
Category
Spreadsheets, databases and docs

Apps and integrations

  • OpenAI
  • Airtable

Node types used

  • Manual Trigger
  • OpenAI Chat Model
  • Airtable
  • Execute Workflow Trigger
  • Edit Fields (Set)
  • HTTP Request
  • Call n8n Workflow Tool
  • If
  • Structured Output Parser
  • OpenAI
  • AI Agent
  • Switch

How it works

The text below comes from the sticky notes inside the workflow. We removed links, code samples and anything that looks like a key or token.

Try It Out!

This workflow does the following

  • Scans an Airtable spreadsheet for rows with product photo images.
  • Uses an AI vision model to attempt to identify the product.
  • Uses an AI Agent to research the product on the internet to enrich the product data.
  • Overwrites our Airtable spreadsheet with the enriched data.

Need Help?

Join the Discord or ask in the Forum!

1. Use Airtable to Capture Survey Photos

Read more about AirTable

To enable this workflow, we need a database where we can retreive the title and photo to analyse and write the generate values back to. Airtable is perfect for this since it has a robust API we can work with.

For this demo, we'll manually trigger but this can be changed for forms or other triggers.

  • Set Airtable Base and Table IDs here.

3. Build an AI Agent who Searches the Internet

Read more about OpenAI Agents

This AI Agent has the ability to perform reverse image searches using our captured photos as well visit external webpages in order to obtain accurate product names and attributes. The Agent along with the tools might mimic what the average human user would carry out the same task.

  • For reverse image search, we're using SERP API service however we won't use the built-in SERP node as we need to specify custom parameters.
  • For scraping, we'll use Firecrawl as this service also helps to parse and return the page as markdown which is more efficient.

4. Overwrite our Rows with Enriched Results

And Viola! Our AI agent has potentially saved hours of manual data entry work for our surveyor. This technique can be used for many other usecases.

5.1 Google Reverse Image Tool

This tool uses Google's reverse image API to return websites where similar images are found.

5.2 Webscraper Tool

This tool uses Firecrawl.dev API to crawl webpages and returns those pages in markdown format.

  • Set Airtable Base and Table IDs here.

Setup notes

The text below comes from the sticky notes inside the workflow. We removed links, code samples and anything that looks like a key or token.

2. Use AI Vision Model to Analyse the Photo.

Read more about OpenAI Vision

We'll use OpenAi vision model to create a detailed description of the product in the photo. We split this step from the agent because it uses an image model rather than the usual text-based one.

5. Using the Custom Workflow Tool

Read more about Workflow Tools

AI Agents rely on Tools to make decisions and become exponentially more powerful the more tools they have. A common pattern to manage multiple tools is to create a routing system for tools using the API pattern.

Nodes in this workflow

The 19 nodes in file order (sticky notes excluded). Branches and connections determine execution order in n8n:

  1. When clicking "Test workflow" (Manual Trigger)
  2. OpenAI Chat Model1 (OpenAI Chat Model)
  3. Get Applicable Rows (Airtable)
  4. Execute Workflow Trigger (Execute Workflow Trigger)
  5. Edit Fields (Edit Fields (Set))
  6. Fallback Response (Edit Fields (Set))
  7. SERP Google Reverse Image API (HTTP Request)
  8. Reverse Image Search Response (Edit Fields (Set))
  9. Reverse Image Search Tool (Call n8n Workflow Tool)
  10. Firecrawl Scrape API (HTTP Request)
  11. Scrape Success? (If)
  12. Firecrawl Scrape Success Response (Edit Fields (Set))
  13. Firecrawl scrape Error Response (Edit Fields (Set))
  14. Firecrawl Web Scaper Tool (Call n8n Workflow Tool)
  15. Structured Output Parser (Structured Output Parser)
  16. Enrich Product Rows (Airtable)
  17. Analyse Image (OpenAI)
  18. Object Identifier Agent (AI Agent)
  19. Actions Router (Switch)

Credentials you need

  • Airtable API – used by Enrich Product Rows and Get Applicable Rows
  • HTTP Header Auth – used by Firecrawl Scrape API
  • OpenAI API – used by Analyse Image and OpenAI Chat Model1
  • Serp API – used by SERP Google Reverse Image API

Configure your own credentials after importing. Create each credential in n8n, then select it in the node.

How to import into n8n

  1. Download the JSON file with the button above.
  2. In n8n, open a new workflow, select the ... menu and choose Import from File. You can also paste the JSON straight into the editor canvas.
  3. Open each node that shows a warning and select or create its credential.
  4. Read the notes on the canvas, test the workflow with sample data, then activate it.

Do you need somewhere to run it? Read our guide to self-hosted n8n.

Need this customised?

We can adapt this workflow to your apps, data and process, then deploy and support it.

Contact Us