Dual-Source Expert Content Generator using Lookio and Linkup - n8n Workflow

• Content marketers and agencies needing factual, source-verified content.


• Technical writers who must leverage proprietary internal knowledge (RAG).


• Automation specialists looking for advanced n8n templates for AI orchestration.


• Teams requiring scalable content generation processes built on the reliable n8n node execution environment.

Generating high-quality, authoritative content typically requires manual research across internal documents and public web sources. This n8n workflow solves this by fully automating the research and writing process, ensuring the final output is grounded in verifiable data.

This specialized n8n template first uses an AI planner to break down the topic into granular research questions. For each question, the n8n node orchestration simultaneously queries an internal knowledge base (Lookio) and performs real-time external web-scraping/search (Linkup). The resulting, structured data is then passed to a powerful LLM (GPT-5) whose sole instruction is to write an article based only on the verified research provided. This sophisticated n8n workflow ensures speed, accuracy, and depth, making it one of the most advanced n8n templates available for RAG-based content creation.

This automation is triggered via a public web form, functioning as a custom n8n trigger. Here is the step-by-step logic:

1. Start & Input: The workflow starts with the 'New article form' n8n trigger, collecting the article title and specific guidelines.


2. Planning Phase: The 'New content - generate research questions' n8n node uses a powerful LLM to decompose the main topic into 5-8 specific sub-questions, structured as a JSON array.


3. Iteration Setup: The array of questions is split out and fed into the 'Loop Over Questions' batch n8n node, initiating a sequential research cycle for each sub-question.


4. Dual-Source Research: Within the loop, two HTTP Request n8n nodes execute:

1. Data Consolidation: The 'Format question and insights' n8n node combines the internal and external results for that specific sub-question.


2. Aggregation: Once all questions are answered, the 'Aggregate full research' n8n node compiles the entire structured Q&A research into one comprehensive payload ('Content to leverage').


3. Final Drafting: The 'New content - Generate the AI output' n8n node takes the aggregated research and the original article brief, using the research as the definitive source material to write the final, fully-sourced article. This entire process is managed reliably by the underlying n8n workflow engine.

To deploy this expert content generation n8n workflow, follow these steps:

1. Import the n8n Workflow: Copy the provided JSON data and import it directly into your n8n instance via the 'Workflows' section.


2. Configure LLM Credentials: Ensure your OpenAI credentials are set up for the 'GPT 5 mini' and 'GPT 5 chat' n8n nodes.


3. Set up Lookio: In the 'Query Lookio Assistant' HTTP Request n8n node, you must replace the placeholders with your Lookio API Key (in the header parameters) and your specific Assistant ID (in the body parameters).


4. Set up Linkup: In the 'Query Linkup for AI web-search' HTTP Request n8n node, connect your Linkup credentials (typically an HTTP Bearer Auth type).


5. Activate the n8n Trigger: Save and activate the workflow. You can then access the public URL provided by the 'New article form' n8n trigger to start generating content.

New article form (n8n trigger): The starting point of the n8n workflow. It uses a Form Trigger to collect two critical inputs: Article Title and Article Guidelines.
New content - generate research questions (Chain LLM n8n node): Function is to act as an AI planning agent. It uses GPT-5 mini to break the large task into small, manageable research questions, ensuring structured JSON output via the 'Structured Output Parser' n8n node.
Split Out Questions (Split Out n8n node): Prepares the research questions for parallel or sequential processing by turning the JSON array into multiple individual items, crucial for the loop structure of this n8n template.
Query Lookio Assistant (HTTP Request n8n node): Retrieves proprietary data. It queries the Lookio API for internal knowledge base insights related to the current research question.
Query Linkup for AI web-search (HTTP Request n8n node): Retrieves external data. It uses Linkup's AI search capabilities to find 5 specific, high-quality, and sourced web insights for the current question.
Aggregate full research (Aggregate n8n node): Collects all researched Q&A pairings back into a single item, creating the comprehensive RAG data payload for the final writing step.

• New content - Generate the AI output (Chain LLM n8n node): The final LLM orchestrator. It uses the combined research to write the complete article, following the original guidelines provided to the n8n workflow.