Chat with GitHub specs using RAG (Pinecone and OpenAI)
Workflow Description
This automation enables direct conversation with GitHub specifications using Retrieval-Augmented Generation (RAG), storing documents in Pinecone and retrieving answers via OpenAI with conversation memory maintained throughout the session.
How it works
- 1.Load GitHub specifications and split them into smaller text chunks
- 2.Store chunks as vectors in Pinecone vector database for semantic search
- 3.Receive user queries through a real-time chat interface
- 4.Search Pinecone for documents most relevant to the query
- 5.Send retrieved context with the question to OpenAI for accurate responses
- 6.Maintain conversation history for coherent multi-turn dialogue
Use cases
- Development teams query API requirements and usage guidelines documented in GitHub specifications
- Project managers quickly search specific details in documentation without manual browsing
- Support teams answer customer inquiries by referencing official documentation instantly
Requirements
- Valid OpenAI API key with sufficient credits
- Pinecone project configured with an active index for vector storage
- Access to GitHub repository containing project specifications
- HTTP endpoint to receive and process user queries
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 (17)
When clicking ‘Test workflow’
Manual Trigger
HTTP Request
HTTP Request
Pinecone Vector Store
Vector Store Pinecone
Default Data Loader
Document Default Data Loader
Recursive Character Text Splitter
Text Splitter Recursive Character Text Splitter
When chat message received
Chat Trigger
AI Agent
Agent
OpenAI Chat Model
OpenAI
Window Buffer Memory
Memory Buffer Window
Vector Store Tool
Tool Vector Store
OpenAI Chat Model1
OpenAI
Sticky Note
Sticky Note
Sticky Note1
Sticky Note
Generate User Query Embedding
OpenAI
Pinecone Vector Store (Querying)
Vector Store Pinecone
Generate Embeddings
OpenAI
Sticky Note2
Sticky Note