Document search via chat with AI and vector search in…

This workflow automates document retrieval using semantic search powered by OpenAI and Supabase. A user asks a question via chat — the system finds relevant files, extracts content, and returns a structured response with source references. Ideal for teams storing documentation in the cloud and seeking fast, accurate access without manual browsing.

## Who it´s for
- Knowledge managers using Supabase for document storage
- Developers building AI agents for file-based search
- Teams automating support with internal documentation
- Owners of knowledge bases with PDF/text files in cloud storage
- SaaS startups needing semantic search over uploaded files

## What the automation does
- Receives a user query via chat
- Converts the query into an embedding using OpenAI
- Performs vector similarity search in Supabase (pgvector) by file metadata and descriptions
- Filters results by relevance threshold and retrieves content snippets
- Generates an AI-powered response with source citations and file links
- Supports manual execution to update embeddings after bulk file uploads

## What´s included
- Ready-to-use n8n workflow
- Trigger logic: chat message received and manual execution
- Integrations with Supabase, OpenAI, and Google Drive
- Basic setup and adaptation guide

## Requirements for setup
- n8n instance with HTTP API access
- Supabase project with pgvector enabled and a table for embeddings
- OpenAI API key for embedding generation and response synthesis
- Google Drive access (optional, if files are stored there)

## Benefits and outcomes
- Reduces document search time to seconds
- Lowers support team workload
- Enables on-demand updates of the vector knowledge base
- Can be embedded into support interfaces or internal portals
- Semantic accuracy beyond keyword matching
- Scalable solution built on open technologies (PostgreSQL + pgvector)

## Important: template only
Important: you are purchasing a ready-made automation workflow template only. Rollout into your infrastructure, connecting specific accounts and services, 1:1 setup help, custom adjustments for non-standard stacks and any consulting support are provided as a separate paid service at an individual rate. To discuss custom work or 1:1 help, contact via Telegram: @gleb923.