ManualTrigger Automate

ManualTrigger Automate streamlines the process of fetching files from Google Drive, splitting them into manageable chunks, and integrating with LangChain for advanced querying. This 22-node workflow allows users to easily interact with documents, generate embeddings, and receive structured responses, enhancing productivity and data accessibility.

7/8/2025
22 nodes
Complex
manualcomplexlangchainsticky notegoogle driveadvanced
Categories:
Complex WorkflowManual Triggered
Integrations:
LangChainSticky NoteGoogle Drive

Target Audience

Target Audience


- Data Scientists: Those who work with large datasets and need to extract insights from documents.
- Researchers: Individuals needing to analyze and query academic papers or reports efficiently.
- Business Analysts: Professionals who require quick access to specific information from various documents.
- Developers: Tech-savvy users looking to integrate AI capabilities into their applications using LangChain and n8n.
- Educators: Teachers or trainers who want to leverage AI for interactive learning experiences using documents.

Problem Solved

Problem Solved


This workflow addresses the challenge of efficiently querying and retrieving information from large documents stored in Google Drive. By utilizing AI embeddings and a vector database, users can extract relevant information quickly, along with citations, thus saving time and enhancing productivity. It enables users to interact with their documents in a conversational manner, making the retrieval process intuitive and effective.

Workflow Steps

Workflow Steps


1. Manual Trigger: The workflow begins when the user clicks the "Execute Workflow" button.
2. Set File URL: The workflow sets the Google Drive file URL that will be processed.
3. Download File: The specified file is downloaded from Google Drive.
4. Add Metadata: Metadata such as file name and URL is added to the downloaded file for future reference.
5. Insert into Vector Store: The file is split into chunks and inserted into a Pinecone vector store for efficient querying.
6. Chat Trigger: Users can initiate a chat to ask questions based on the document content.
7. Get Top Chunks: The workflow retrieves the top relevant chunks from the vector store based on the user's query.
8. Prepare Chunks: The retrieved chunks are prepared for processing.
9. Generate Embeddings: AI embeddings are created for the prepared chunks to facilitate understanding.
10. Answer Query: The AI generates a response to the user's query using the context from the document chunks.
11. Compose Citations: Citations are compiled to provide references for the generated answer.
12. Generate Final Response: The final response, including the answer and citations, is prepared for the user.

Customization Guide

Customization Guide


- Change File Source: Users can modify the Google Drive file URL in the "Set File URL in Google Drive" node to point to different documents.
- Adjust Chunk Size: In the "Recursive Character Text Splitter" node, users can change the chunkSize and chunkOverlap parameters to control how the document is split.
- Modify Query Parameters: Users can adjust the topK value in the "Get Top Chunks Matching Query" node to retrieve more or fewer chunks based on their needs.
- Customize AI Models: Users can replace the OpenAI model credentials with their own to use different AI capabilities.
- Add Additional Processing Steps: Additional nodes can be added to enhance the workflow, such as data transformation or integration with other APIs.