Knowledge engine to build personalized and dynamic AI memory for agents.
Author: topoteretes
Version: 0.0.2
Type: tool
This is an AI memory plugin for Dify. Cognee is an open-source knowledge engine that lets you ingest data in any format or structure and continuously learns to provide the right context for AI agents. It combines vector search, graph databases and cognitive science approaches to make your documents both searchable by meaning and connected by relationships as they change and evolve.
Create a new dataset or return an existing one with the same name. Use this before ingesting data if you want to set up the dataset explicitly.
Returns: ,
Ingest text content into a dataset. Multiple text items can be separated by newlines.
Returns: , , ,
Upload files into a dataset. Accepts documents, images, and other file types supported by Cognee. Files are passed as variables from chat input or workflow start nodes.
Returns: , ,
Transform ingested data into a knowledge engine. This runs a multi-step pipeline: classifying documents, extracting text chunks, identifying entities and relationships via LLM, generating summaries, and embedding everything into vector and graph stores.
This is a long-running operation — processing time depends on data volume.
Returns:
Search within the cognee memory using one of 14 search strategies. Each strategy is suited to different use cases.
Read more about the search types in cognee docs: link
Returns: ,
List all datasets accessible to the authenticated user. No parameters required.
Returns: ,
List all data items stored in a specific dataset.
Returns: ,
Permanently delete a dataset and all its associated data.
Returns: ,
Delete a specific data item from a dataset.
Returns: , ,