app icon
cognee
0.0.2

Knowledge engine to build personalized and dynamic AI memory for agents.

topoteretes/cognee314 installs

Cognee

Author: topoteretes
Version: 0.0.2
Type: tool

Description

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.

Setup

  1. Get your API key and base URL from your Cognee Cloud dashboard.
  2. Install the plugin in your Dify workspace.
  3. Configure it with your Base URL (e.g. ) and API Key.

Tools

ToolPurpose
Create DatasetCreate a named dataset to hold your data
Add DataIngest text into a dataset
Add FileUpload files (PDF, DOCX, TXT, etc.) into a dataset
CognifyProcess a dataset into a searchable knowledge engine
SearchQuery the knowledge engine
Get DatasetsList all available datasets
Get Dataset DataList all data items in a dataset
Delete DatasetPermanently delete a dataset
Delete DataDelete a specific data item from a dataset

Create Dataset

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.

ParameterRequiredDescription
Dataset NameyesName for the new dataset

Returns: ,


Add Data

Ingest text content into a dataset. Multiple text items can be separated by newlines.

ParameterRequiredDescription
Text DatayesText content to add (newline-separated for multiple items)
Dataset NamenoTarget dataset name (either this or Dataset ID required)
Dataset IDnoTarget dataset UUID (either this or Dataset Name required)
Node SetnoComma-separated node set names for graph organization

Returns: , , ,


Add File

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.

ParameterRequiredDescription
FilesyesFiles to upload (variable from chat or workflow)
Dataset NamenoTarget dataset name (either this or Dataset ID required)
Dataset IDnoTarget dataset UUID (either this or Dataset Name required)
Node SetnoComma-separated node set names for graph organization

Returns: , ,


Cognify

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.

ParameterRequiredDescription
DatasetsnoComma-separated dataset names (either this or Dataset IDs required)
Dataset IDsnoComma-separated dataset UUIDs (either this or Datasets required)
Custom PromptnoCustom prompt for entity extraction and graph generation
Ontology KeynoComma-separated keys referencing previously uploaded ontology files

Returns:


Search

Search within the cognee memory using one of 14 search strategies. Each strategy is suited to different use cases.

ParameterRequiredDefaultDescription
QueryyesNatural language search query
Search TypeyesSearch strategy (see table below)
DatasetsnoComma-separated dataset names to search
Dataset IDsnoComma-separated dataset UUIDs to search
System PromptnoCustom system prompt for completion-type searches
Node NamenoComma-separated node set names to filter results
Top Kno10Maximum number of results
Only ContextnofalseReturn raw retrieval context instead of LLM completion
VerbosenofalseInclude additional details in the response

Read more about the search types in cognee docs: link

Returns: ,


Get Datasets

List all datasets accessible to the authenticated user. No parameters required.

Returns: ,


Get Dataset Data

List all data items stored in a specific dataset.

ParameterRequiredDescription
Dataset IDyesUUID of the dataset to inspect

Returns: ,


Delete Dataset

Permanently delete a dataset and all its associated data.

ParameterRequiredDescription
Dataset IDyesUUID of the dataset to delete

Returns: ,


Delete Data

Delete a specific data item from a dataset.

ParameterRequiredDescription
Dataset IDyesUUID of the dataset
Data IDyesUUID of the data item to delete

Returns: , ,


Typical Workflow

  1. Create Dataset — set up a named dataset.
  2. Add Data or Add File — ingest text or documents into the dataset.
  3. Cognify — build the knowledge engine from the ingested data.
  4. Search — search within cognee memory
  5. Get Datasets / Get Dataset Data — inspect what's available.
  6. Delete Data / Delete Dataset — clean up when needed.

Links

CATEGORY
Tool
VERSION
0.0.2
topoteretes·04/11/2026 09:32 AM
REQUIREMENTS
Tool invocation
Maximum memory
256MB