Sync WordPress & WooCommerce to Pinecone with OpenAI Embeddings (n8n)

Keep your Pinecone search fresh with WordPress + WooCommerce content synced via OpenAI embeddings (n8n)

This n8n workflow automatically syncs your WordPress pages and blog posts (via WooCommerce/WordPress REST API endpoints) into a Pinecone vector index using OpenAI embeddings—updating changed content and removing deleted items.

What this workflow does

• Runs on demand or on a schedule: manual trigger or daily cron.
• Fetches published content in batches: pulls up to 100 published WordPress pages and up to 100 published posts (IDs and modified timestamps) via the REST API.
• Detects new or changed items: compares live IDs/modified dates against the indexed_wc_blog_v1 n8n Data Table, then selects up to staticBatchSize items that need indexing.
• Loads full content: retrieves the complete page/post data for the selected batch, combines results, and strips HTML before processing.
• Embeds + indexes into Pinecone: for each page/post, it prepares metadata, deletes any existing Pinecone vectors for the same fileId, splits text into chunks, generates OpenAI embeddings, and inserts chunk vectors into your Pinecone index.
• Keeps track of what’s indexed: upserts the page/post fileId and dateModified into indexed_wc_blog_v1.
• Deletes stale vectors: finds entries present in indexed_wc_blog_v1 but missing from WordPress, deletes their Pinecone vectors, and removes their rows from the Data Table.

Use cases

• Power semantic search or RAG over your WordPress site content with a Pinecone-backed index.
• Ensure WooCommerce/WP changes propagate automatically to your vector database after publishing or updates.
• Prevent “ghost” results by automatically removing deleted or unpublished pages/posts from Pinecone.

Technical details

• Integrations: WordPress/WooCommerce REST API, OpenAI embeddings API, Pinecone.
• n8n nodes / building blocks: set, code, no op, data table, sticky note, http request.
• State management: uses the indexed_wc_blog_v1 n8n Data Table to track indexed IDs and dateModified.