# oh-my-hermes

An operating layer installed on top of Hermes Agent that adds planning, evidence-backed workflows, and honest reporting to coding, research, and operations tasks.

oh-my-hermes (OMH) is an operating layer that sits above Hermes Agent, an existing AI agent, without replacing it. It turns a plain natural-language request into a defined capability, a concrete next step, and an honest record of what actually happened, adding explicit evidence boundaries to planning, research, content creation, coding handoffs, operations, and project memory.

It installs with a shell script on macOS and Linux or a PowerShell script on Windows, after which a setup command configures it and a model picker lets the user assign different models to different categories of work by effort level. An update command detects how it was installed and upgrades itself along with its managed skills and Hermes registration, and a doctor command verifies the installation.

It is written in Python and distributed under the MIT license.

Website: https://rlaope.github.io/oh-my-hermes/