# cognee

> MCP Server

Cognee is a memory layer for AI apps and agents, enabling dynamic agent memory using scalable ECL pipelines to interconnect and retrieve past conversations, documents, images, and audio transcriptions for

## Overview

- **Category:** AI
- **Language:** Jupyter Notebook
- **Stars:** 29494
- **Forks:** 137
- **Owner:** topoteretes
- **GitHub:** https://github.com/topoteretes/cognee
- **Homepage:** https://www.cognee.ai
- **Created:** 2023-08-16T16:16:33+00:00
- **Updated:** 2025-03-28T23:23:43+00:00
- **Source:** https://model-context-protocol.com/servers/llm-memory-ai-agents-conversations-documents

## Setup

## Setup

```
import os
os.environ["LLM_API_KEY"] = "YOUR OPENAI_API_KEY"

```

You can also set the variables by creating .env file, using our <a href="https://github.com/topoteretes/cognee/blob/main/.env.template">template.</a>
To use different LLM providers, for more info check out our <a href="https://docs.cognee.ai">documentation</a>

## Tools

## Available Tools

1. Interconnect and retrieve your past conversations, documents, images and audio transcriptions (allows connecting and retrieving various data types).
2. Reduce hallucinations, developer effort, and cost (aims to improve AI response reliability and efficiency).
3. Load data to graph and vector databases using only Pydantic (simplifies data loading to databases).
4. Manipulate your data while ingesting from 30+ data sources (enables data manipulation during ingestion from multiple sources).


## Related Articles

- [cognee: An MCP Server for AI Agent Memory via ECL Pipelines](https://model-context-protocol.com/blog/cognee-an-mcp-server-for-ai-agent-memory-via-ecl-pipelines-mcp-server-guide)