AI-Driven LinkedIn Post Generator with Human Approval - n8n Workflow

• Content creators and marketing specialists managing corporate social media profiles.


• Consultants who follow specific, high-engagement content strategies (like the 'Jeu Entrepreneurial').


• Users looking for advanced examples of human-in-the-loop validation in their n8n templates.


• Professionals needing a reliable, automated pipeline for turning documents (like meeting notes or articles) into social media updates.

Generating consistent, high-quality content that adheres to specific marketing frameworks is challenging. This specialized n8n workflow solves this by streamlining the entire process. It accepts various inputs (form submission with files, email content) and intelligently routes them for processing. The core value lies in the custom AI Agent n8n node, which is meticulously programmed (using French, entrepreneurial marketing principles) to generate highly personal and engaging LinkedIn posts from raw text, adhering to a 4-step conversion-focused methodology. By integrating an explicit Send Email n8n node for human approval, this automation ensures no content is published until validated, merging the speed of automation with necessary quality control. This is a robust example of a complex, multi-stage n8n workflow.

This automation begins with one of two possible n8n trigger points: either an On form submission trigger, allowing for file uploads (PDF, DOCX) or raw text input, or an Email Trigger (IMAP) monitoring specific inbound emails.

1. Input Routing: The Switch n8n node analyzes the input MIME type. Documents are routed to the appropriate Extract from File n8n node (for PDF or TXT) or through an intermediate step leading to a Google Docs n8n node for fetching content from document files.


2. AI Transformation: The extracted text is fed into the AI Agent n8n node. This agent, utilizing an OpenAI Chat Model (GPT-4.1 mini), applies a rigorous, pre-defined 4-step French LinkedIn strategy (focused on 'Le jeu entrepreneurial') to transform the source material into an optimized post.


3. Human-in-the-Loop: The resulting post text is immediately sent to a designated email address via the Send Email n8n node, which pauses the n8n workflow and awaits a response through a custom validation form (requiring a 'oui' or 'non' answer).


4. Approval and Publishing: The subsequent Switch1 n8n node checks the email response. If the validation is 'oui', the n8n workflow proceeds to the final LinkedIn n8n node, publishing the AI-generated content publicly. If 'non', the automation terminates.

To deploy this powerful n8n workflow, follow these setup steps:

1. Import: Copy the provided n8n workflow JSON and paste it into your n8n instance using the 'New' -> 'Import from JSON' option.


2. OpenAI Credential: Locate the OpenAI Chat Model n8n node and set up your OpenAI API key credentials. This is crucial for the AI Agent to function.


3. LinkedIn Credential: Configure the LinkedIn n8n node with your OAuth2 credentials. Ensure the account linked has permission to publish posts.


4. Email Credentials: Update the Send Email n8n node with valid SMTP credentials for sending the approval request and the IMAP credentials for the Email Trigger (IMAP) if you intend to use that input method.


5. Form/Recipient Setup: In the Send Email n8n node, modify the toEmail field to your personal approval address. If using the On form submission n8n trigger, share the generated webhook URL to access your input form.


6. Activate: Save the n8n workflow and switch the workflow to 'Active' to start listening for inputs.

On form submission (n8n trigger): Initiates the n8n workflow when a custom form is submitted, allowing users to upload files (fichier) or paste text.
Switch (n8n node): Essential for flow control. Determines the file type (PDF, DOCX, TXT) of the uploaded document to ensure correct subsequent extraction logic.
Extract from File (n8n node): Handles the physical content extraction from binary files, configured specifically for PDF content.
Google Docs (n8n node): Used to retrieve content from document links, typically following a document upload/conversion process.
OpenAI Chat Model (n8n node): Provides the language model (GPT-4.1-mini) for the AI Agent. Requires the OpenAI API credential.
AI Agent (n8n node): The core intelligence. It uses an extensive system message in French detailing a professional LinkedIn marketing strategy (entrepreneurial game) to guide content generation from the input text.
Send Email (n8n node): Acts as the human approval gate. It sends the AI-generated post to the user and pauses the n8n workflow execution until a custom 'oui/non' form response is received.
Switch1 (n8n node): Checks the approval response from the email form. Only if the user responds 'oui' does the n8n workflow continue to publishing.

• LinkedIn (n8n node): Publishes the final, validated text output from the AI Agent as a public LinkedIn post.