AI agents often lose context across sessions. By integrating Notion with Cognee, agents can automatically ingest workspace documentation and convert it into an interconnected Knowledge Graph.
How it works
I developed NotionLoader using the official notion-client Python SDK. It fetches pages and databases, extracts text content, and passes it to Cognee's ingestion pipeline.
Python
from notion_loader import NotionLoader
import cognee
loader = NotionLoader()
pages = loader.load_data()
for page in pages:
cognee.add(f"Title: {page['title']}\n\n{page['content']}")
cognee.cognify()
This brings seamless knowledge graph memory to any Cognee agent working with Notion documents!