Turn one uploaded image into a clean product catalog in Airtable. Great for marketing and ops teams that collect vendor or competitor logos and want fast, structured data without manual typing.

A form collects a logo sheet image and an optional hint. An AI agent reads the image and returns a structured list of tools, their attributes, and similar items. A parser enforces a clear JSON format. The flow splits each tool and attribute, checks Airtable for existing records, and creates missing ones. A crypto step builds stable hashes from names to avoid duplicates. Code steps convert attribute names and similar tools into Airtable record links so everything connects.

You need an Airtable base with Tools and Attributes tables, plus an OpenAI API key. Expect large time savings and cleaner data, because new and existing records are handled in one run. This fits competitor tracking from event photos, quick product research, and catalog updates from marketing files. After setup, open the form URL, upload an image, and watch Airtable fill with linked records.

• Form upload trigger collects a logo sheet image and an optional hint.
• AI agent reads the image and extracts tools, attributes, and similar items into a clear schema.
• Structured Output Parser ensures valid JSON before any database write.
• Split nodes loop through tools and attributes for reliable batch handling.
• Airtable upsert creates attributes if missing and reuses them when found.
• Crypto step makes a stable hash from the tool name for consistent IDs.
• Code steps map attribute names and similar tools to Airtable record IDs.
• Linked records connect each tool to attributes and to similar tools.
• Merges and batches coordinate timing so records exist before links are added.

• Reduce manual data entry from hours to minutes by parsing one image into many records.
• Streamline catalog updates by upserting records and avoiding duplicates.
• Improve data quality by enforcing a structured output and mapped record links.
• Handle higher volume by batching tools and attributes without extra staff time.
• Connect AI extraction with Airtable to keep one source of truth.

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Airtable and OpenAI. See the Tools Required section above for links to create accounts with these services.
3. In Airtable, create a base with two tables. Tools: Name, Attributes linked to Attributes table, Hash, Similar linked to Tools, optional Description, Website, Category. Attributes: Name, Tools linked to Tools.
4. In your Airtable account, create a Personal Access Token with access to the base and required scopes. Keep the token safe.
5. In the n8n credentials manager, create a new Airtable Personal Access Token credential. Double click any Airtable node, choose Create new credential, and paste your token. Select the correct base and tables in node options.
6. In your OpenAI account, create an API key. In n8n, create a new OpenAI credential and paste the key. On the gpt node, select the gpt 4o model.
7. Open the Retrieve and Parser Agent node and review the system message and the structured schema. Adjust the prompt to match your use case.
8. Open the Form Trigger node. Keep or change the path. Note the form URL shown in the node. Confirm the file upload field is required.
9. Run a test. Open the form URL, upload a logo sheet image, and submit. In the n8n execution, check the Structured Output Parser, Split Out Tools, and All Attributes nodes for expected values.
10. In Airtable, confirm that attributes exist and that each tool links to its attributes and to similar tools. If links are empty, verify field types are linked records to the correct tables.
11. If you see duplicates, check the Airtable upsert settings and make sure the Name fields match exactly. If extraction is thin, add context in the Additional Prompt field and submit again.