What is Hermes Agent, and what does an open-source team platform add?
Hermes Agent is the self-hosted personal AI agent built by Nous Research. It remembers across sessions, writes its own skills, and reaches you through messaging apps. When a whole team needs governed agents, Kortix is the open-source AI Operating System that puts agents, skills, memory, and connectors in one git repo you own.
What Kortix is
Kortix is the open-source AI Operating System.Your agents, skills, company memory, and connectors live in one git repo you own.Each session runs on its own cloud computer.The definition and what it runs on, from the Kortix documentation.
What Hermes Agent is
Hermes Agent is an open-source, self-hosted AI agent built by Nous Research and released in February 2026. It runs on hardware you control and reaches you through a single messaging gateway, so you can start a conversation in Telegram and continue it in your terminal. Nous Research describes it as an agent that lives on your server, remembers what it learns, and gets more capable the longer it runs (hermes-agent.org).
Hermes Agent keeps persistent memory across sessions, writes reusable skill documents when it solves a hard problem, and works with any model provider you point it at. Its repository lists several terminal backends for where the agent runs, including local, Docker, SSH, and hosted sandboxes, and it installs with one shell command on Linux, macOS, and WSL2 (github.com/nousresearch/hermes-agent).
Who makes Hermes Agent
Nous Research builds Hermes Agent and publishes it under its own name. The project site and the repository both credit the company, and the source install path is a single curl command. The messaging gateway is the part that makes it feel personal. It connects Telegram, Discord, Slack, WhatsApp, Signal, and email from one process, and the agent follows you between them.
Is Hermes Agent self-hosted?
Yes. Hermes Agent installs on hardware you control, and its memory, skills, and configuration stay there. Nous Research's repository says it runs on a $5 VPS, a GPU cluster, or serverless infrastructure that costs nearly nothing when idle (github.com/nousresearch/hermes-agent). The vendor charges no seat. That model keeps the agent and its accumulated context on your side, and it puts the model bill and the compute bill in your hands.
Where a single-operator agent stops fitting a team
Hermes Agent is designed around one operator, and the project's own language treats it as a personal agent. A team that adopts it runs into questions the design does not answer: who may run which tool, what each session may reach, and what record exists of what an agent did. Those are governance questions, and they grow with the size of the team.
Kortix is the open-source AI Operating System: your agents, their skills, your company memory, and every connector in one git repo you own, with the agents working on real cloud computers. It carries the pieces a team needs around a personal agent. Per-resource permissions for people and agents. Roles, groups, and an audit trail. One isolated sandbox per session. Each session has its own isolated machine and branch. Connector policy sets each tool to Allow, Ask or Block, down to the arguments of each call.
Read the Hermes alternatives page for the tools teams compare before they switch, and the Kortix vs Hermes comparison for the two side by side.
Hermes Agent questions
- Which apps can I reach Hermes Agent from?
- Hermes Agent runs a single gateway process that connects Telegram, Discord, Slack, WhatsApp, Signal, and email. You can start a session in a messaging app and continue it in the terminal (github.com/nousresearch/hermes-agent).
- Does Hermes Agent run on Windows?
- The project ships a native PowerShell install, and its own documentation describes native Windows as experimental and recommends WSL2 for a stable setup (github.com/nousresearch/hermes-agent).
Put a whole team’s agents in one repo you own.
Start on Kortix Cloud, or self-host for free.
Open source · Any model, your keys · Self-host, VPC, or on-prem