Logo
Deploy Now

Categories

Self-Hosted Developer Tools Machine Learning Vector Database Agent Infrastructure Knowledge Graph AI Memory

Developer links

Last Updated: 01-01-2023 12:00 AM (UTC)

Cognee

Cognee gives AI agents persistent long-term memory that survives across sessions, replacing the traditional stack of separate graph, vector, and session databases with a unified engine running on a single PostgreSQL instance. The memory-native API exposes four verbs (remember, recall, forget, and improve) enabling agents to persist context, retrieve cited answers, prune outdated knowledge, and self-improve from feedback. Under the hood, Cognee combines pgvector embeddings with a PostgreSQL-native graph store and cognitive-science-grounded ontology generation, delivering hybrid retrieval that fuses semantic similarity, structural graph traversal, and lexical search in a single query. Integrations span Claude Code, Cursor, LangGraph, OpenAI Agents, and any MCP-compatible client through a dedicated MCP server on port 8001, while the Python and TypeScript SDKs provide direct programmatic access. The platform supports swappable backends including Neo4j, FalkorDB, Qdrant, ChromaDB, Weaviate, Milvus, and LanceDB for teams with existing infrastructure. Built-in OpenTelemetry tracing, an experimental dashboard with knowledge graph visualization, multi-tenant user isolation, and audit trails ensure production readiness. Deploy via Docker Compose with optional profiles for PostgreSQL, Neo4j, Redis, and the web frontend. Reached v1.0 in April 2026 with 30,000+ stars. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

Cognee
Cognee
Cognee
Cognee
Cognee

Benefits

  • Unified Memory on Single Database
  • Run graph, vector, and relational storage on one PostgreSQL instance with pgvector, eliminating the need to deploy and manage separate Neo4j, Redis, and vector database services.
  • Four-Verb Memory API
  • The remember, recall, forget, and improve verbs provide session-aware persistence with automatic background sync from fast session cache to permanent knowledge graph storage.
  • Universal Agent Integration
  • Connect any AI agent through the built-in MCP server, Python SDK, TypeScript SDK, or first-party plugins for Claude Code, Cursor, LangGraph, and OpenAI Agents.
  • Self-Improving Knowledge Graph
  • The improve function enables agents to learn from feedback, corrections, and interaction patterns, evolving memory relevance and relationship quality over time without manual curation.

Features

  • Hybrid Retrieval Engine
  • Combines semantic vector similarity, structural graph traversal, and lexical search in a single query against PostgreSQL with pgvector and graph store.
  • MCP Server
  • Exposes memory tools on port 8001 for IDE integrations with Claude Code, Cursor, and any Model Context Protocol compatible AI client.
  • OpenTelemetry Tracing
  • Native OTEL instrumentation across all API functions and pipeline runs provides observability with any compatible collector like Jaeger or Datadog.
  • Multi-Tenant Isolation
  • Per-user and per-tenant data separation with authentication, access control, and audit trails for production deployments serving multiple customers.
  • Swappable Backends
  • Replace default PostgreSQL storage with Neo4j, FalkorDB, Qdrant, ChromaDB, Weaviate, Milvus, or LanceDB through community adapters and configuration.