# Website Knowledge Chatbot (RAG + Supabase)

**Category:** AI Chatbot  
**Tools:** OpenAI, HTTP, Supabase  
**Nodes:** 18  
**Built by:** Forgent Systems

Turns any website into a chatbot that answers from the site's own content. Pages are scraped, embedded and stored in Supabase, and questions are answered with retrieved context.

![Workflow](preview.webp)

## How it works

1. Enter a website URL to start ingestion.
2. The page content is scraped, cleaned and split into chunks.
3. Embeddings are created and stored in the Supabase vector store.
4. Chat questions retrieve the most relevant chunks and AI writes the answer.

## Use cases

- Website support bots
- Documentation assistants
- Internal knowledge bases

## Setup

1. In n8n open **Workflows → Import from File** and choose `workflow.json`.
2. Connect these credentials:
   - Cohere account
   - OpenAI account
   - Postgres account
   - Supabase account
3. Point the Sheets / Airtable / Drive / Slack nodes at your own resources (structure above).
4. Replace placeholder emails, phone numbers and `YOUR-FORM-ID` links with your own.
5. Times use the America/New_York timezone and dates use MM/DD/YYYY. Adjust if needed.
6. Run a test execution, then activate.

## Nodes

- **When chat message received** — `@n8n/n8n-nodes-langchain.chatTrigger`
- **HTML Extract** — `n8n-nodes-base.htmlExtract`
- **Default Data Loader** — `@n8n/n8n-nodes-langchain.documentDefaultDataLoader`
- **Recursive Character Text Splitter** — `@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter`
- **Supabase Vector Store** — `@n8n/n8n-nodes-langchain.vectorStoreSupabase`
- **Enter Website Url** — `n8n-nodes-base.formTrigger`
- **Sticky Note1** — `n8n-nodes-base.stickyNote`
- **Sticky Note2** — `n8n-nodes-base.stickyNote`
- **Sticky Note3** — `n8n-nodes-base.stickyNote`
- **Sticky Note** — `n8n-nodes-base.stickyNote`
- **Website Data Scraping** — `n8n-nodes-base.httpRequest`
- **Convert to File** — `n8n-nodes-base.convertToFile`
- **Embeddings Cohere** — `@n8n/n8n-nodes-langchain.embeddingsCohere`
- **Question & Answer Retrieve** — `@n8n/n8n-nodes-langchain.agent`
- **Data From Supabase Vector Store** — `@n8n/n8n-nodes-langchain.vectorStoreSupabase`
- **Embeddings With Cohere** — `@n8n/n8n-nodes-langchain.embeddingsCohere`
- **Chat Memory** — `@n8n/n8n-nodes-langchain.memoryPostgresChat`
- **OpenAI Chat Model** — `@n8n/n8n-nodes-langchain.lmChatOpenAi`
