Host your own AI deep research agent with n8n, Apify and OpenAI o3

by Jimleuk ✓ · 79,829 views · created 2025-02-11 · Market Research, AI Chatbot

What this workflow does

This AI workflow has 64 nodes and runs when someone submits an n8n form (“ DeepResearcher”) (plus 1 other trigger). Along the way it connects OpenAI Chat Model, Notion and Google Gemini Chat Model. It also calls api.apify.com over HTTP.

How it runs

Trigger: When someone submits an n8n form (“ DeepResearcher”) (plus 1 other trigger).

1. Research Request (n8n Form): shows the next form page
2. Execution Data: uses Execution Data
3. Set Variables (Edit Fields (Set)): sets 4 fields: request_id, prompt, depth…
4. JobType Router (Switch): routes items into 3 branches
5. Clarifying Questions (Basic LLM Chain): sends a prompt to an LLM, with model: OpenAI Chat Model; output parser: Structured Output Parser
6. Get Existing Row (Notion): get many (database page)
7. Generate SERP Queries (Basic LLM Chain): sends a prompt to an LLM, with model: OpenAI Chat Model; output parser: Structured Output Parser
8. Get Existing Row (Notion): get many (database page)
9. Feedback to Items (Split Out): splits a list into separate items
10. Set In-Progress (Notion): update (database page)

…plus 41 more steps. See the full node list below.

Services it uses

OpenAI Chat ModelNotionGoogle Gemini Chat Model

HTTP calls to: api.apify.com

Before you upgrade to n8n 3.0

Nodes used (23)

• Structured Output Parser@n8n/n8n-nodes-langchain.outputParserStructured
• Edit Fields (Set)n8n-nodes-base.set
• OpenAI Chat Model@n8n/n8n-nodes-langchain.lmChatOpenAi
• n8n Form Triggern8n-nodes-base.formTrigger
• Basic LLM Chain@n8n/n8n-nodes-langchain.chainLlm
• Split Outn8n-nodes-base.splitOut
• No Operation, do nothingn8n-nodes-base.noOp
• Loop Over Items (Split in Batches)n8n-nodes-base.splitInBatches
• n8n Formn8n-nodes-base.form
• Execute Workflow Triggern8n-nodes-base.executeWorkflowTrigger
• Execute Sub-workflown8n-nodes-base.executeWorkflow
• Execution Datan8n-nodes-base.executionData
• Switchn8n-nodes-base.switch
• Notionn8n-nodes-base.notion
• Ifn8n-nodes-base.if
• Markdownn8n-nodes-base.markdown
• Google Gemini Chat Model@n8n/n8n-nodes-langchain.lmChatGoogleGemini
• HTTP Requestn8n-nodes-base.httpRequest
• Filtern8n-nodes-base.filter
• Mergen8n-nodes-base.merge
• Coden8n-nodes-base.code
• Aggregaten8n-nodes-base.aggregate
• Stop and Errorn8n-nodes-base.stopAndError

From the author

This template attempts to replicate OpenAI's DeepResearch feature which, at time of writing, is only available to their pro subscribers. > An agent that uses reasoning to synthesize large amount of online information and complete multi-step research tasks…

View the original on n8n.io → Check your own copy