topoteretes/cognee

Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine.

What it solves

Cognee provides AI agents with persistent, long-term memory across different sessions. It solves the problem of agents forgetting project context, past decisions, and learned rules between runs, allowing them to maintain a "company brain" of shared knowledge from documentation, code, and conversations.

How it works

Cognee transforms raw data—such as text and code—into a structured memory system consisting of entities, relationships, and searchable chunks. It uses a hybrid approach where text becomes a knowledge graph and code is mapped as a graph of symbols and dependencies.

The system supports two types of memory: permanent memory (stored in a knowledge graph) and session memory (a fast cache that can be distilled into permanent memory). At query time, an auto-routing mechanism selects the best retrieval strategy (graph, vector, or code context) to provide the agent with the relevant evidence needed to answer a question or perform a task.

Who it’s for

  • AI Agent Developers: Those building agents that need to remember context across sessions or maintain domain-specific knowledge.
  • Enterprise Teams: Organizations wanting to create a shared "Company Brain" from their internal documentation and tickets.
  • Software Engineers: Developers integrating memory layers into Python, TypeScript, Rust, or HTTP-based applications.

Highlights

  • Multi-format Ingestion: Supports text, code, and session-based lessons.
  • Hybrid Memory: Combines knowledge graphs, vector embeddings, and symbol graphs for code.
  • Session Distillation: Curates useful lessons from temporary sessions into durable, long-term memory.
  • Broad Integration: Offers plugins for Claude Code, Codex, and MCP clients (Cursor, Cline), as well as SDKs for multiple languages.
  • Flexible Deployment: Can be self-hosted via Docker or used as a managed cloud service, with an experimental option to run the entire memory layer on a Postgres instance.

Related

  • Dispatch
    Bolnee-Chat: Self-Hosted RAG Chatbot for Business Websites

    Bolnee-Chat is an open-source, self-hosted RAG chatbot platform that allows businesses to integrate grounded AI assistants into their websites using a two-line embed snippet.

  • Project
    24kchengYe/MemoMind

    MemoMind is a locally hosted, GPU‑accelerated memory system for AI coding agents. It stores extracted facts from chats, documents, and daily life events in PostgreSQL + pgvector, builds a knowledge graph, and offers fast 4‑way hybrid retrieval (semantic, BM25, graph, temporal). The agent can retain new information, recall relevant memories, and reflect across the whole store. All data stays on the user’s machine, works with any OpenAI‑compatible LLM, and includes a web dashboard for browsing and exporting memories.

  • Project
    doobidoo/mcp-memory-service

    A persistent, shared memory backend for AI agents that provides a knowledge graph and semantic search across sessions and tools via REST API and MCP.

  • Project
    mnemon-dev/mnemon

    A persistent, LLM-supervised memory layer for AI agents that enables cross-session knowledge retention across multiple AI frameworks using a four-graph knowledge store.

  • Project
    superlinked/sie

    A self-hosted inference engine that serves multiple open-source models for agent tasks—such as retrieval, OCR, and generation—through a single, unified API.