repowiserepowise

Command-line tool

rlaope / oh-my-hermes

⚚ The ultrawork agent with AI coworkers wired into you

★ 5,749stars on GitHubPythonMIT

This connects Repowise to rlaope/oh-my-hermes's code. It does not install oh-my-hermes.

Repowise maps public repos from their source. Not affiliated with rlaope. What is Repowise?

github.com/rlaope/oh-my-hermesIndexed at cda327a

What is oh-my-hermes?

oh-my-hermes is an operating layer for Hermes Agent, used to run several coworker agents in parallel while you keep talking to Hermes. It places a separate agent with its own skill and model on each lane and does not patch or replace Hermes. The project is written mainly in Python, and its OMHER Desktop app uses TypeScript and Rust. To install it, run brew install rlaope/tap/omh and follow the steps below. Its code health is 6.4 out of 10 (Fair), ranked #180 of 4,000+ indexed repos. The last commit to oh-my-hermes was on Nov 10, 2026, and 43 people have contributed.

As of commit cda327a, indexed Nov 11, 2026. What it is comes from a model reading the README; the rest is measured from the source and git history, not from the maintainer.

For agentsoh-my-hermes.md

Repo facts: type, license, health, rank, contributors
oh-my-hermes facts
TypeCommand-line tool
LicenseMIT
Code health6.4 / 10 (Fair)
Health rank#180 of 4,000+ indexed repos
Last commitNov 10, 2026
Contributors43
Bus factor1
Size3,899 files, 1,032,186 lines

Install oh-my-hermes

Commands copied from the README

  1. 1Install the omh command with Homebrew.
    brew install rlaope/tap/omh
  2. 2Or install the omh command with npm, which also needs Node 18+ and Python 3.11+ on PATH.
    npm install -g oh-my-hermes
  3. 3Run the setup after installing.
    omh setup

Releases on GitHub (latest v5.0.0, Nov 21, 2026)

Full install notes on GitHub

Explore oh-my-hermes

Ask about this code in your agent

This connects Repowise to oh-my-hermes's code. It does not install oh-my-hermes. To install oh-my-hermes, use the install steps.

Useful if you want to change or fork oh-my-hermes: your agent can ask how it is put together and get answers that cite its files.

How to use oh-my-hermes

  1. 1Verify the installation or troubleshoot problems.
    omh doctor

Ask about oh-my-hermes

Answers come from oh-my-hermes's README and source, with the files they cite. Two free answers a day without an account.

Featured answer

oh-my-hermes is an operating layer for Hermes Agent that runs several coworker agents in parallel while you keep talking to Hermes, without patching or replacing Hermes. Each lane gets a separate agent with its own skill and model, and input enters through the omh command, where bare omh opens the project's terminal interface on Hermes' gateway. The src directory holds most of the Python code, including src/tui/app.py for the interface and src/skills/render.py and src/skills/catalog_definitions.py for the skills catalog that the router loads into runs as tool calls. The OMHER Desktop app in apps/omher is written in TypeScript and Rust, starting from apps/omher/src/app/App.tsx for the React entry point and apps/omher/src-tauri/src/main.rs for the Rust shell. For reading order, the Application layer is the largest with 2036 of 3899 files, so begin with src, then the tests directory, and .claude/manifest.json with .claude/install.sh for the managed setup.

Files per layer in oh-my-hermes8 layers.

Basis: Inferred · knowledge graph· as of cda327a

Want this for your own codebase? Map your repo free

About this page

Get a page like this for your own codebase

Repowise built this map of rlaope/oh-my-hermes from its source and git history. It parses every symbol, draws the dependency graph, scores each file's health from complexity, duplication, coverage and churn, finds hotspots and owners in the history, and writes documentation for every module. Your coding agent reads the same index over MCP, so its answers cite real files.

Free for 2 public repos, with a page like this one. No card. See how Repowise works

Prefer local? pip install repowise && repowise init indexes your own checkout, no account needed. Star Repowise on GitHub

rlaope/oh-my-hermes FAQ

What is oh-my-hermes?

oh-my-hermes is an operating layer for Hermes Agent that puts a coworker with its own skill and model on each lane, in parallel. You keep talking to Hermes, and OMH never patches or replaces it.

How does oh-my-hermes work?

oh-my-hermes turns a goal given through /dag-ulw into a graph of small nodes, and the ready nodes run in parallel, each on its category's model and effort. A separate node checks the goal before the report is given.

How do I install oh-my-hermes?

Install oh-my-hermes with brew install rlaope/tap/omh or npm install -g oh-my-hermes, which provide the omh command. Then run omh setup, and use omh doctor to verify or troubleshoot the installation. See the install steps

What language is oh-my-hermes written in?

oh-my-hermes is written mainly in Python, which accounts for 2617 of its 3899 files. Its OMHER Desktop app uses TypeScript and Rust, and the repository also includes JavaScript and shell files.

How is the oh-my-hermes code organized?

oh-my-hermes keeps its core code in src and its test suite in tests, with the desktop app in apps/omher. Skills are stored in the skills and agent-skills directories.

When should I use oh-my-hermes?

Use oh-my-hermes when you want Hermes Agent to run several tasks at once, each on its own lane with its own model and skill. It counts a run as verified only when a test, review or CI gate has passed.

What license does oh-my-hermes use?

oh-my-hermes is released under the MIT license, identified as MIT. The license applies to the whole repository.

Code health

6.4out of 10Fair

Hotspots average 4.9, the place fixes pay off most.

On the leaderboards:#50 of 67 Python repos

Key features of oh-my-hermes

  • Each lane is tagged with a work category that picks its model chain and reasoning effort, and the chain moves to the next model when a provider rejects one.
  • A goal given through /dag-ulw becomes a graph of small nodes, and the ready nodes run in parallel, each on its category's model.
  • Expert skills are loaded into a run by the router as tool calls, so they are not called by hand.
  • When LSP tools are enabled, definitions, references and renames come from the language server rather than text search.
  • Each memory candidate is kept, refused or deferred with a written reason, and the next session gets a recall pack cut to a token budget.
  • Every status names its stage, and a run counts as verified only when a test, review or CI gate actually passed.

How rlaope/oh-my-hermes works

Input enters through the omh command, and bare omh opens OMH's own terminal interface on Hermes' gateway. The src directory holds most of the Python code, and src/tui/app.py is one of its hotspot files for that interface. src/skills/render.py and src/skills/catalog_definitions.py cover the skills catalog, which the router loads into a run as tool calls. The desktop app lives in apps/omher, where apps/omher/src/app/App.tsx is the React entry point and apps/omher/src-tauri/src/main.rs starts the Rust shell. The tests directory holds the test suite, and .claude/manifest.json with .claude/install.sh are central files for the managed setup.

The codebase has 3,899 files and 76,460 symbols in 19 modules, led by Python, TypeScript and JavaScript. Code health is 6.4 out of 10, rated fair.