AI Agents

Chat with GitHub specs using RAG (Pinecone and OpenAI)

17 nodes 213 116 Automatic trigger
Download

Workflow Description

An intelligent automation system that enables interactive conversations with GitHub repository specifications using Retrieval-Augmented Generation (RAG). The system combines Pinecone vector database with OpenAI's language capabilities to deliver precise answers to technical inquiries about project documentation.

How it works

  1. 1.Load GitHub specification documents and split them into logical text chunks
  2. 2.Convert text segments into vector embeddings and store them in Pinecone
  3. 3.Receive user queries through the chat interface and retrieve relevant documents
  4. 4.Use an intelligent agent combining memory and vector search to compose contextual responses
  5. 5.Deliver the final answer to the user while maintaining conversation history

Use cases

  • Answer technical questions about repository structure and GitHub documentation
  • Accelerate information retrieval from extensive project documentation repositories
  • Provide development teams with an intelligent assistant for common project-related questions

Requirements

  • Valid OpenAI API key to run the language model and embedding functions
  • Active Pinecone account with an initialized index for vector storage and retrieval
  • GitHub specification files in text format (Markdown or Plaintext) ready for processing

Service Value

Ideal as a smart automation service combining integrations and AI to produce ready-to-use results.

Apps Used

Vector Store Document Default Data Loader Text Splitter Chat Trigger AI Agent OpenAI Memory

Details

Trigger Automatic trigger
Nodes 17
Apps 7
Views 213
Downloads 116

How to Use

  1. 1.Click "Download Template"
  2. 2.Open your n8n dashboard
  3. 3.Go to Workflows > Import from File
  4. 4.Select downloaded file and configure credentials

Nodes Used (17)

/

When clicking ‘Test workflow’

Manual Trigger

#1

HTTP Request

HTTP Request

#2

Pinecone Vector Store

Vector Store Pinecone

#3

Default Data Loader

Document Default Data Loader

#4

Recursive Character Text Splitter

Text Splitter Recursive Character Text Splitter

#5

When chat message received

Chat Trigger

#6

AI Agent

Agent

#7

OpenAI Chat Model

OpenAI

#8

Window Buffer Memory

Memory Buffer Window

#9

Vector Store Tool

Tool Vector Store

#10

OpenAI Chat Model1

OpenAI

#11

Sticky Note

Sticky Note

#12

Sticky Note1

Sticky Note

#13

Generate User Query Embedding

OpenAI

#14

Pinecone Vector Store (Querying)

Vector Store Pinecone

#15

Generate Embeddings

OpenAI

#16

Sticky Note2

Sticky Note

#17