Memori: Give AI Agents Persistent Memory
Summary
Memori adds structured, persistent memory to LLM applications by capturing agent execution and conversations. Its Python and TypeScript SDKs integrate with existing models and data infrastructure, with managed cloud and BYODB options.
At a glance
- Language
- Python
- License
- NOASSERTION
- Stars
- 17.1k
- Forks
- 3.6k
- Added to OSRepos
- December 28, 2025
- Last analyzed
- October 3, 2026
Use at your own risk
OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of code from these repositories is the user's own responsibility. Always review the repository, source code, dependencies, licenses, and security implications before running or installing anything. OSRepos is not responsible for issues, damages, or losses resulting from third-party repositories.
Overview
Memori is a memory layer for AI agents and LLM applications. It turns conversations and agent execution into structured, persistent state, helping agents retain useful context across sessions rather than repeatedly rediscovering information.
It is designed for teams that want to add memory without replacing their existing model, framework, or data infrastructure. The project offers Python and TypeScript SDKs, integrations for agent tools, a managed cloud service, and a bring-your-own-database option.
Key Features
- Captures conversations and agent execution for later recall.
- Organizes memory by entity, process, and session, with advanced augmentation for facts, preferences, events, and relationships.
- Integrates with Python and TypeScript applications and supports several LLM providers and frameworks.
- Provides plugins and memory providers for OpenClaw and Hermes Agent.
- Offers an MCP connection for compatible coding agents and clients.
- Supports managed Memori Cloud and connecting your own database through BYODB.
- Includes a dashboard for memories, analytics, and API keys, plus a CLI for account and quota tasks.
Use Cases
- Add continuity to customer-support agents that need to recall user preferences across conversations.
- Help coding agents retain project conventions and developer preferences across sessions.
- Reduce repeated investigation in multi-step automation by retaining prior tool calls, decisions, and outcomes.
- Give teams a shared memory layer for production agents while retaining their existing infrastructure.
Project Facts
- Language: Python
- License: NOASSERTION
- Stars: 17.1k
- Forks: 3.6k
- Topics: agent, agent-memory, agenticai, ai, ai-memory, claude-code, enterprise, hermes, llm, long-short-term-memory, memory, memory-management, openclaw, python, rag, state-management, stateful, typescript
- Archived: false
Getting Started
Install the Python SDK with pip install memori or the TypeScript SDK with npm install @memorilabs/memori. The cloud quickstart requires a Memori API key and an LLM API key. See the README for setup and integration details.
Alternatives
- ReMe: ReMe stores evolving, searchable Markdown memory, while Memori captures structured memory from agent execution and conversations through SDKs.
- memoripy: Memoripy is a local, auditable memory runtime with versioned records; Memori offers Python and TypeScript SDKs plus managed cloud and BYODB options.
Considerations
- Memory creation depends on attribution: the README says interactions without an entity and process attribution cannot produce memories.
- The documented cloud quickstart requires API keys and is subject to quota limits. Advanced Augmentation is described as rate-limited without an account.
- BYODB and on-premises deployment are presented as options, but database setup and deployment details require consulting the project documentation.
- The repository lists its license as
NOASSERTION, while its README states Apache 2.0. Verify the license files and terms before adopting it.
Source repository
Open the original repository on GitHub.
24 counted GitHub visits
Related repositories
Similar repositories that may be relevant next.
web-design: A Claude Code SKILL for Spec-First Web Page Design
October 3, 2026
The web-design project is a Claude Code SKILL designed to streamline the creation of beautiful and consistent web pages. It emphasizes a 'spec first, code second' approach, ensuring design principles are established before development begins. This tool helps generate UI, visuals, motion, and responsiveness that are consistent across pages and easily editable.
OOMWOO: Build Your Own Open-Source, Hackable Robot Vacuum Cleaner
October 2, 2026
OOMWOO is an ambitious open-source project enabling users to build their own robot vacuum cleaner using Raspberry Pi, 3D printing, and ROS2. It emphasizes local operation, hackability, and integration with Home Assistant, providing a high-quality, customizable home appliance. This project aims to deliver a fully open hardware, software, and firmware solution for autonomous home cleaning.
Shepherd: Reversible Execution Traces for Programmable Meta-Agents
October 2, 2026
Shepherd is a Python runtime substrate designed for agent work requiring inspection, reversibility, and supervision. It records agent runs as durable, inspectable execution traces, enabling meta-agents to observe, fork, replay, and revert any operation. This framework couples agents and environments using a copy-on-write fork, offering significant performance benefits and robust permission enforcement.
Agent Anvil: CI-First Evaluation Harness for Tool-Using AI Agents
October 1, 2026
Agent Anvil is a robust, CI-first evaluation harness designed for AI agents that utilize tools. It meticulously runs scenario suites, captures detailed traces of agent behavior, and provides semantic grading to identify issues. The platform excels at clustering failures and suggesting concrete fixes for prompts, tools, and guardrails, ensuring agents behave safely and effectively.