Slack AI Chatbot with RAG for Company Staff

This n8n workflow creates an intelligent Slack AI chatbot that leverages Retrieval-Augmented Generation (RAG) to provide accurate, context-aware responses to staff inquiries about company policies, documents, and procedures. It integrates seamlessly with Slack, Google Drive, and a vector database (Qdrant) to enable real-time information retrieval and conversational AI interactions.

The workflow begins with a Slack trigger that detects when the bot is mentioned in a channel. It then captures the message and retrieves relevant company documents by searching a Qdrant vector database, which stores embedded representations of internal documents. The documents are fetched from Google Drive, processed, and converted into embeddings using OpenAI’s API. The AI agent, powered by LangChain, uses these embeddings along with a large language model (Claude or Anthropic) to generate precise, well-formatted responses.

Memory buffer nodes maintain conversation context within a sliding window, enabling the AI to refer back to recent interactions. The chatbot responses are then sent back to Slack, maintaining a threaded conversation for clarity. Additional nodes facilitate the creation and refreshment of the Qdrant collection, supporting dynamic updates to the document repository.

This workflow is ideal for companies aiming to streamline internal knowledge sharing, reduce response times for common questions, and enhance employee productivity through an intelligent, always-available virtual assistant.

Practical use cases include automating responses to HR queries, IT support requests, and internal documentation navigation, improving overall operational efficiency.