Syncing Google Docs to Your AI Knowledge Base

How do I connect Google Docs to ChatNexus?

Direct answer: Go to your ChatNexus agent → <strong>Knowledge Base</strong> → <strong>Add Source</strong> → select <strong>Google Docs</strong> → authorize via OAuth → select which Docs to sync → save.

ChatNexus uses Google's OAuth flow, so you grant permission securely without sharing your password. You can select individual docs or entire folders. After authorization, the sync starts automatically on your chosen schedule (daily, weekly, etc.).

What happens to my Google Docs after I sync them?

Direct answer: ChatNexus reads your docs, chunks them into 1500-character segments with 10% overlap, and stores the chunks in your Neon PostgreSQL database with pgvector embeddings. The original Google Docs remain unchanged.

The chunking process preserves context by overlapping segments—the last 150 characters of one chunk are the first 150 characters of the next. This helps the vector search find relevant information across chunk boundaries. The docs themselves are never modified.

How often are my Google Docs synced?

Direct answer: You choose the sync interval: every hour, daily, weekly, or on-demand. ChatNexus runs background jobs to fetch updates and re-embed changed chunks.

When you update a Google Doc, ChatNexus detects the change and re-chunks only the modified sections. Old embeddings are replaced with new ones. This keeps your knowledge base fresh without re-processing the entire doc every time.

What data storage backend does ChatNexus use?

Direct answer: ChatNexus uses Neon-hosted PostgreSQL with pgvector and pg_trgm extensions. This is a cloud-hosted, highly available database with vector search built in.

Neon provides auto-scaling and serverless PostgreSQL. The pgvector extension enables semantic search via cosine similarity on your embeddings. All data is encrypted at rest and in transit.

Can I sync multiple Google Docs or folders?

Direct answer: Yes. You can add multiple Docs, folders, or even shared Drive folders to a single knowledge base. Each document is chunked and embedded independently.

When the user asks a question, ChatNexus searches across all synced documents, retrieves the top 3 most relevant chunks, and injects them into the system prompt. This is called Retrieval-Augmented Generation (RAG) and ensures accurate, sourced answers.

Deploy an AI agent on your site in minutes

Connect your knowledge base, embed on any domain, and start free — no credit card required.

Create Your AI Agent Free