Local Document Processing and AI Reporting System - n8n Workflow

Legal Professionals & Researchers: Who need automated summarization and extraction from large volumes of local documents (like contracts or case files).
Developers & AI Enthusiasts: Interested in creating highly secure, on-premise AI automation using the n8n platform alongside models like Ollama.
Data Engineers: Needing a resilient n8n workflow trigger mechanism for local file events combined with complex AI processing.
Compliance Officers: Requiring documentation processing that remains entirely within a local environment for security reasons.

This sophisticated n8n workflow template solves the challenge of processing sensitive documents while ensuring data sovereignty. By integrating the n8n Local File Trigger with an advanced AI Agent powered by a locally hosted LLM (often Ollama or Llama 3 configured via an OpenAI compatibility layer), organizations can automate the analysis of legal documents, technical manuals, or financial reports. The value lies in 100% local execution: files never leave your environment, yet you gain powerful automated reporting capabilities. Utilizing this comprehensive n8n node setup ensures seamless file handling from detection to final report generation.

The execution of this robust n8n workflow begins with the Local File Trigger.

1. File Detection (n8n trigger): The n8n workflow starts instantly when a new file is detected in the specified local folder path configured in the n8n trigger node.


2. File Read: The subsequent Read/Write Files from Disk n8n node accesses the newly triggered file from the local filesystem.


3. Content Extraction: The Extract from File n8n node is crucial, designed to intelligently pull readable text content (e.g., OCR or text layer extraction) from documents like PDFs or Word files, preparing the data for the AI.


4. AI Analysis: The extracted text is passed to the AI Agent n8n node. This agent, leveraging the capabilities of the connected Language Model, analyzes the document based on its internal prompt and generates a structured output (e.g., a summarized legal analysis or key finding report).


5. Model Interaction: The OpenAI Chat Model n8n node provides the computational backbone for the agent. In this specific n8n workflow, this connection is typically configured to point to a local Ollama instance running Llama models for high-performance, private AI processing.


6. Report Generation: The AI Agent’s output is converted into a structured file format (like an HTML or Markdown report) using the Convert to File n8n node.


7. Final Write: The report file is written back to a specified output folder on the local disk using the final Read/Write Files from Disk1 n8n node, completing the automated cycle. This entire n8n workflow demonstrates a secure, end-to-end automation solution.

To use this powerful n8n workflow template, follow these steps:

1. Import: Copy the provided JSON definition and import it directly into your self-hosted n8n instance.


2. Local AI Setup (Crucial): Ensure you have a local Large Language Model (LLM) running, such as Ollama. This LLM must be exposed via an OpenAI-compatible API endpoint (e.g., using a reverse proxy or Ollama's native API).


3. Credential Setup:

1. Activation: Once paths and credentials are set, activate the n8n workflow to begin monitoring the specified input folder.

This specialized n8n workflow utilizes several key nodes for local processing and AI interaction:

Local File Trigger (n8n trigger):
Function: Serves as the starting point, initiating the n8n workflow when a file appears in the designated input directory.
Key Configuration: Requires specification of the monitored folder path and the file extension pattern to match.
Read/Write Files from Disk (n8n node):
Function: Reads the detected file from the local disk, providing the binary data to the next steps of the n8n workflow.
Key Configuration: Operation set to 'Read'. Requires correct file path configuration.
Extract from File (n8n node):
Function: Extracts raw text content from the binary input file payload, making the data readable for the subsequent AI step. Essential for processing documents like PDFs.
Key Configuration: Automatically processes the incoming binary data fields.
OpenAI Chat Model (n8n node):
Function: Acts as the interface to the Language Model. Although labeled OpenAI, it is configured here to connect the AI Agent to a local model endpoint (like Ollama or Llama 3).
Key Configuration: Base URL override is critical for pointing to the local LLM instance.
AI Agent (n8n node):
Function: Orchestrates the analysis. It takes the extracted text, applies specific reasoning and prompts, and utilizes the connected LLM to produce a structured, actionable result within this n8n workflow.
Key Configuration: Contains the prompt instructions defining the analysis task (e.g., "Summarize the legal obligations in the following text and output as HTML").
Convert to File (n8n node):
Function: Packages the text output from the AI Agent (the generated report) back into a binary file format, ready to be saved to disk.
Key Configuration: Defines the output filename and desired format (e.g., report.html).
Read/Write Files from Disk1 (n8n node):
Function: The final step, used to handle direct file system write operations, saving the generated report back to a local output directory. Completes the n8n workflow loop.
* Key Configuration: Operation set to 'Write'. Requires the output file path and name.