Memori: memory tool for Hermes Agent

LLM-agnostic memory layer with Python and TypeScript SDKs that persists and recalls agent conversations

What Memori does

Memori registers with your LLM client through its Python or TypeScript SDK. You set an attribution of entity and process, and Memori persists conversations in the background and recalls them in later requests. The README says it works with the data infrastructure you already run and deploys across managed cloud, single-tenant cloud, VPC and on-premises.

Memori Cloud needs only an API key, a bring-your-own-database option is documented, and a dashboard offers Memories, Analytics, Playground and API Keys. The project reports 87% accuracy on the LoCoMo benchmark with an average of 721 tokens per query, which it describes as 2.8% of the full-context footprint; these are its own figures. A README section covers a plugin for OpenClaw gateways that captures structured memory, including tool calls, decisions and outcomes, after each turn. The opening part of the README does not describe a Hermes integration.

Key features

Python and TypeScript SDKs that register with an existing LLM client
Conversations persisted and recalled automatically in the background
Attribution by entity and process, such as a user and an agent
Plugin for OpenClaw gateways that captures tool calls, decisions and outcomes
Dashboard with Memories, Analytics, Playground and API Keys
Bring-your-own-database option and VPC or on-premises deployment

When to use it

Adding persistent memory to an app built on an OpenAI client
Giving OpenClaw agents memory between sessions through the plugin
Deploying agent memory inside a VPC or on-premises for enterprise systems

Who it is for: Developers building production agent systems who want a hosted or self-managed memory layer behind their LLM calls.

How it fits with Hermes Agent

The repository carries the hermes topic, but the opening part of the README documents SDK use and an OpenClaw plugin rather than a Hermes integration. Hermes Agent users should check the Memori docs for current support.

How to install Memori

These commands are copied from the project's README. Check the repository for the latest steps before you run them.

npm install @memorilabs/memori
pip install memori

Requirements: A Memori API key (MEMORI_API_KEY) and an LLM API key such as OPENAI_API_KEY

FAQ

What is Memori?

Memori is agent-native memory infrastructure: an LLM-agnostic layer that turns agent execution and conversation into structured, persistent state. It ships Python and TypeScript SDKs and a hosted cloud.

Does Memori work with Hermes Agent?

The repository is tagged hermes, but the opening part of the README documents an OpenClaw plugin and SDK use, not a Hermes integration. Check the Memori docs before relying on Hermes support.

How do I install Memori?

Install the TypeScript SDK with npm install @memorilabs/memori or the Python SDK with pip install memori. Then set MEMORI_API_KEY and your LLM API key and register your client.