Chat with PDF Documents Using AI with Source Citations
Workflow Description
Automated workflow for interactive conversations with PDF content using advanced language models, automatically extracting and citing sources from the original text to ensure accuracy and traceability of information.
How it works
- 1.Load PDF files from Google Drive or local file system
- 2.Split text into processable segments and generate embeddings for semantic search
- 3.Process user queries through chat interface and retrieve relevant text passages
- 4.Send relevant passages to language model to generate cited responses
Use cases
- Answer questions about contract and report contents while maintaining document references
- Analyze legal or medical documents and extract relevant information with proper attribution
Requirements
- Valid OpenAI API key with appropriate permissions
- PDF documents stored in Google Drive or accessible local file system
Service Value
Ideal as a smart automation service combining integrations and AI to produce ready-to-use results.
Apps Used
Details
How to Use
- 1.Click "Download Template"
- 2.Open your n8n dashboard
- 3.Go to Workflows > Import from File
- 4.Select downloaded file and configure credentials
Nodes Used (22)
When clicking "Execute Workflow"
Manual Trigger
Embeddings OpenAI
OpenAI
Sticky Note
Sticky Note
Default Data Loader
Document Default Data Loader
Set file URL in Google Drive
Set
Sticky Note2
Sticky Note
Add in metadata
Code
Download file
Google Drive
Chat Trigger
Chat Trigger
Prepare chunks
Code
Embeddings OpenAI2
OpenAI
OpenAI Chat Model
OpenAI
Set max chunks to send to model
Set
Structured Output Parser
Output Parser Structured
Compose citations
Set
Generate response
Set
Sticky Note1
Sticky Note
Answer the query based on chunks
LLM Chain
Sticky Note4
Sticky Note
Get top chunks matching query
Vector Store Pinecone
Add to Pinecone vector store
Vector Store Pinecone
Recursive Character Text Splitter
Text Splitter Recursive Character Text Splitter