AI-powered Customer Knowledge Base Assistant!

Workflow 1: Website Scraping & Data Storage:
This workflow automates the process of collecting and preparing knowledge for an AI support agent:

1. Trigger: Starts when clicking "Execute workflow."
2. Fetch Links: Retrieves site links from a Google Sheet.
3. Loop & Scrape: Iterates through each link and scrapes the website data.
4. Vectorization: Converts scraped content into embeddings using OpenAI embeddings.
5. Data Storage: Stores the vectorized data in Pinecone vector database for semantic search.
6. Update Status: Marks the link as processed in the Google Sheet. This ensures that all customer-facing knowledge is continuously updated and accessible in a structured format.

Workflow 2: AI-Powered Customer Support Agent:
This workflow uses the stored knowledge to answer queries in real time and integrates with multiple tools for enhanced support:

1. Trigger: Starts when a chat message is received.
2. AI Agent: Uses the stored embeddings in Pinecone to understand the query and retrieve the most relevant information.
3. Context & Memory: Maintains conversational memory with Simple Memory and uses the OpenAI Chat Model for natural responses.
4. Actions: o Fetch user query data from Pinecone for context. o Send meeting details via email. o Add events to calendar for client meetings. o Send Slack notifications to the team. This makes the support agent not just a responder, but a proactive assistant that can handle scheduling, notifications, and knowledge-based answers seamlessly.

With these workflows combined, we have:
✔ Automated web scraping and knowledge management
✔ AI-powered customer support with contextual memory
✔ Multi-channel integrations (Email, Calendar, Slack)
✔ Scalable and efficient workflow automation