AI Agents

AI agent for conversing with files stored in Supabase

33 nodes 308 181 Automatic trigger
Download

Workflow Description

An automation enabling direct conversation with files stored in Supabase using an AI agent. The system loads files, splits them into chunks, generates semantic embeddings via OpenAI, stores them in a vector database, then answers user queries based on file contents through intelligent retrieval.

How it works

  1. 1.Load files from Supabase storage and extract text content
  2. 2.Split document text into meaningful chunks for processing
  3. 3.Generate semantic embeddings using OpenAI's embedding model
  4. 4.Store embeddings in Supabase vector database for retrieval
  5. 5.Receive user questions through the chat interface
  6. 6.Search vector database and provide contextual answers

Use cases

  • Answer questions about stored PDFs, presentations, and documents
  • Analyze company reports and contracts interactively
  • Extract specific information from large document collections

Requirements

  • Active Supabase account with configured vector table
  • Valid OpenAI API key for embeddings and chat models
  • Support for file formats and defined file size limits

Service Value

Ideal as a smart automation service combining integrations and AI to produce ready-to-use results.

Apps Used

Document Default Data Loader Text Splitter OpenAI Supabase Split In Batches Chat Trigger Vector Store AI Agent

Details

Trigger Automatic trigger
Nodes 33
Apps 8
Views 308
Downloads 181

How to Use

  1. 1.Click "Download Template"
  2. 2.Open your n8n dashboard
  3. 3.Go to Workflows > Import from File
  4. 4.Select downloaded file and configure credentials

Nodes Used (33)

/

Get All files

HTTP Request

#1

Default Data Loader

Document Default Data Loader

#2

Recursive Character Text Splitter

Text Splitter Recursive Character Text Splitter

#3

Extract Document PDF

Extract From File

#4

Embeddings OpenAI

OpenAI

#5

Create File record2

Supabase

#6

If

If

#7

Get All Files

Supabase

#8

Download

HTTP Request

#9

Loop Over Items

Split In Batches

#10

When clicking ‘Test workflow’

Manual Trigger

#11

Aggregate

Aggregate

#12

When chat message received

Chat Trigger

#13

OpenAI Chat Model1

OpenAI

#14

Embeddings OpenAI2

OpenAI

#15

OpenAI Chat Model2

OpenAI

#16

Vector Store Tool1

Tool Vector Store

#17

Switch

Switch

#18

Insert into Supabase Vectorstore

Vector Store Supabase

#19

Merge

Merge

#20

AI Agent

Agent

#21

Supabase Vector Store

Vector Store Supabase

#22

Sticky Note

Sticky Note

#23

Sticky Note1

Sticky Note

#24

Sticky Note2

Sticky Note

#25

Sticky Note3

Sticky Note

#26

Sticky Note4

Sticky Note

#27

Sticky Note5

Sticky Note

#28

Sticky Note9

Sticky Note

#29

Sticky Note7

Sticky Note

#30

Sticky Note6

Sticky Note

#31

Sticky Note8

Sticky Note

#32

Sticky Note10

Sticky Note

#33

Similar Templates

Extract Google Drive files, analyze with AI, and populate Google Sheets

34 nodes

Analyze Google Sheets data and send intelligent Slack notifications

12 nodes

Read and process emails using artificial intelligence

11 nodes

Query Google Sheets data intelligently through Slack

12 nodes

Extract Gmail data and save to Google Sheets using AI analysis

24 nodes

Build an intelligent agent that processes documents and answers queries

21 nodes