Universal-memory provides a vendor-agnostic persistence layer for AI agents, enabling consistent memory across different LLM providers. It supports advanced context engineering, agent skills, and integrates with the Model Context Protocol (MCP). Aimed at developers building reliable, multi-provider autonomous AI systems.
universal-memory is an AI & ML project. It focuses on giving AI agents consistent, provider-independent long-term memory and cognitive state management. It is built as an open-source project for AI developers. The project is open source (Apache-2.0). It ships for the web, the command line, and API.
Behind universal-memory is YanAmorelli, and it first shipped in 2026. The project is developed in the open on GitHub with 226 commits in the last 90 days. Among its 4 catalogued features are Cognitive Persistence, Vendor Agnostic, and Agent Memory. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do