OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Hermes Fleet Memory

Hermes Fleet Memory is a Python project that provides shared vector memory and MCP-based retrieval for multiple Hermes Agent instances across machines. It uses Qdrant for vector storage and documents standalone and distributed deployment paths, with domain configuration for separating work, personal, shared, or all-domain memory.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Hermes Fleet Memory — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when separate agent instances need to retrieve shared vector memory, or when you are configuring an MCP memory server and can operate the required local or self-hosted infrastructure.

Who it is for

  • Developers configuring persistent, searchable context for multiple Hermes Agent instances.
  • Operators connecting agents running on different computers to a shared Qdrant-backed memory service.
  • Users who want to configure an MCP memory server for Hermes Agent or Claude Desktop.

Documented uses

  • Set up a standalone Fleet Memory node and run its documented preflight check.
  • Configure an MCP server entry for an AI client.
  • Choose a standalone or distributed deployment path for shared vector retrieval.
  • Set a domain value for the node's memory configuration.
Before you choose

Requirements & limits

The repository is public and carries an MIT license. Its README describes deployments on infrastructure the operator owns rather than a shared hosted Fleet Memory instance.

  • The documented stack includes Python 3.10+, FastMCP, and Qdrant.
  • The standalone quickstart differs from distributed deployments, which require selecting and configuring a Qdrant endpoint.
  • The project provides client configuration examples for Hermes Agent and Claude Desktop, and a registration sequence for Claude Code.
  • Domain values documented by the project include work, personal, shared, and all.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up a standalone node and run its preflight check

Clone the repository, initialize standalone mode, and inspect the node with the documented doctor command.

Before you start

  • Python 3.10 or later, as indicated by the repository
  • A local environment where the repository and its setup script can be run

Steps

  1. Clone the repository and change into its directory.

  2. Run the command below.

    Documented command
    python setup.py --role standalone --non-interactive

  3. Run the command below.

    Documented command
    python client/fleet_memory.py --doctor

Expected outcome

The standalone setup flow has been invoked and the node's bounded preflight check has been run.

Guide limitations

  • This sequence does not configure an AI client or demonstrate retrieval.
  • The repository describes the doctor command as a check, not as a deployment or execution test.
Documentation 1 ↗
Source-based practical guide

Configure the MCP client for Hermes Agent

Add the documented Fleet Memory server entry to Hermes Agent configuration.

Before you start

  • A local clone of the repository
  • Hermes Agent configuration access
  • Python 3 and the absolute path to client/fleet_memory.py

Steps

  1. Open ~/.hermes/config.yaml.

  2. Add an mcp_servers entry named fleet-memory with command python3 and the absolute script path as its argument.

  3. Set the FLEET_HARD_DOMAIN environment value to the intended domain; the README's example uses all.

Expected outcome

The documented Hermes Agent MCP server configuration is in place.

Guide limitations

  • The configuration example alone does not establish that the server is connected or that retrieval works.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
amrlazw/hermes-fleet-memory · Official homepage title · retained in its source language
How we review sourcesSuggest a correction