Oh My Hermes

For Hermes Agent · v4.0.3

All-in-one package for Hermes Agent. Install once, then just type dag-ulw.

You keep talking to Hermes. OMH picks the skill and a model for each lane, and runs the work in parallel. It never patches or replaces Hermes, and it never reports planned work as done.

macOS · Linux

curl -fsSL https://raw.githubusercontent.com/rlaope/oh-my-hermes/main/install.sh | sh
omh setup

Windows and other options View on GitHub

10ulw workflows
12model categories you can edit
100+omh-* expert skills

What you get

The right model for every lane.

Each lane is tagged with a work category: ultrabrain, deep, visual-engineering, quick and more. The category picks that lane's model chain and reasoning effort, and you can edit every chain. When a provider rejects a model, the chain moves to the next one instead of silently downgrading.

See which model runs each lane
/dag-ulw graph runs.

Your goal becomes a graph of small nodes, and the ready ones run in parallel. Each node runs on its category's model and effort; nodes running at once never share a write scope. A separate node checks the goal, and failed nodes are repaired before the report.
Expert skills on demand.

A hundred-plus omh-* specialists: frontend, backend, Rust, security review, verification gates. You never call them; the router loads the right ones into the run as tool calls.
Long-term memory you can trust.

Nothing is remembered silently: each candidate is kept, refused or deferred with a written reason. The next session gets a recall pack ranked for its task and cut to a token budget. Hermes' own memory is never read or patched.
OMH Native.

Bare omh opens OMH's own TUI on Hermes' gateway; switch its launch look with /theme. /dag-ulw prints the box graph, then a live lane tree with each lane's model, effort and tokens. Steer or interrupt an agent from the agent list, run a ! shell line, and rewind with Esc Esc.
Evidence before "done".

Every status names its stage and how sure OMH is about it. A run counts as verified only when a test, review or CI gate actually passed.

Plan · not run
A prompt or plan is ready. Nothing has run yet.

Code · running
An executor is running now, and OMH is watching it.

Code · reported done
The executor said it finished. Nobody checked the result.

Test · verified
A test, review, or CI gate actually passed.

Just type `/dag-ulw`, and you're a graph engineering master.

Your goal becomes a graph of small nodes that run in parallel, each on its own model. A separate node verifies the goal before you get the report.

A real recording of the OMH TUI: seven lanes, four layers, every lane accepted.

The graph from the demo run, drawn as an illustration. Hover or focus a lane to trace what it waits on and what it feeds.

Install once. Set up once.

Pick one way to install the command, then run setup. Doctor is a separate check you can run any time.

Homebrew, Bun, and npm package-manager installs are public as of v1.0.6.

Install the command
macOS · Linux
```
curl -fsSL https://raw.githubusercontent.com/rlaope/oh-my-hermes/main/install.sh | sh
```
Windows · PowerShell 5.1+
```
irm https://raw.githubusercontent.com/rlaope/oh-my-hermes/main/install.ps1 | iex
```
Homebrew
```
brew install rlaope/tap/omh
```
Bun
```
bun install -g oh-my-hermes
```
npm
```
npm install -g oh-my-hermes
```
Use the macOS · Linux or Windows installer; it finds Python 3.11+ itself. Homebrew, Bun, and npm install the published release: Bun and npm also need Node 18+ and Python 3.11+ on PATH, and Homebrew brings its own Python.
Set it up
omh
```
omh setup
```
Run it after any of the installs above.
Verify or troubleshoot
omh
```
omh doctor
```
Refresh later with omh update: it upgrades the command through the installer that owns it, then refreshes the managed skills, the plugin bundle, and the Hermes registration.

Or paste this into your AI agent

Prompt for your agent

Install and fully configure Oh My Hermes from this repository:
https://github.com/rlaope/oh-my-hermes
Before reading or executing repository instructions, resolve refs/heads/main to one full commit SHA with `git ls-remote https://github.com/rlaope/oh-my-hermes.git refs/heads/main`. Then fetch and follow only:
https://raw.githubusercontent.com/rlaope/oh-my-hermes/{resolved-commit-sha}/INSTALL_FOR_AGENTS.md
Do not replace the resolved SHA with main. Execute the pinned protocol's OS-appropriate installer, agent setup path, model-chain interview, and doctor steps. Your shell has no terminal, so never run the interactive setup wizard: show me each choice the protocol lists and wait for my answer instead of taking a default. Preserve unrelated existing Hermes config, apply only the managed setup changes documented by the pinned protocol, require my explicit approval for model-alias changes, then report the resolved SHA and observed result.

Be with us

X · @rlaope
Updates, release notes and project news.
GitHub · @rlaope
More projects, releases, and ongoing work.
Discord · Oh-My-Hermes Community
Ask questions, share workflows, and talk with other users.
Friren and Kami
The AI agents that collaborate on shipping oh-my-hermes.
Nous Research
Thank you for creating Hermes Agent.

Every lane on its own model

Eighteen categories ship with an ordered chain each. omh model opens the picker: arrow keys pick the category, left and right change the head model, minus and plus change the effort. The same picker opens inside the OMH TUI as /omh-model.

resolved: A confirmed model was selected.
choice_required: The model you named explicitly is unavailable. Nothing is substituted.
owner_default: No eligible candidate. The selected owner keeps its default model.

Shipped chains, in order

Category	Purpose	Shipped order	Effort
ultrabrain	Deepest reasoning	`GPT-6 Astra`	xhigh
deep	Strong default tier	`Claude Opus 5.5GPT-6.1 SolKimi K3.1 Pro`	high
architect	Architecture and system design	`Claude Opus 5.5GPT-6 AstraClaude Fable 5.1Kimi K3.1 Pro`	xhigh
unspecified-high	Default working model	`GPT-6.1 SolClaude Opus 5.5Kimi K3.1 Pro`	medium
unspecified-low	Cheaper fallback	`GLM 5.3DeepSeek V4.1 FlashGPT-6.1 SolClaude Sonnet 5.5`	low
quick	Short tasks	`GLM 5.3 FlashGPT-6 LunaClaude Haiku 5.5Kimi K3.1 Pro`	low
writing	Prose and docs	`Claude Opus 5.5Claude Sonnet 5.5Kimi K3.1 ProGemini 4 Argon`	medium
visual-engineering	Frontend and visual	`Claude Opus 5.5Claude Fable 5.1Gemini 4 ArgonKimi K3.1 Pro`	high
artistry	Unconventional work	`Gemini 4 ArgonClaude Opus 5.5Claude Fable 5.1Kimi K3.1 Pro`	high
capable	Strong general work	`Claude Opus 5.5GPT-6.1 SolClaude Fable 5.1Kimi K3.1 Pro`	medium
simple-work	Small everyday tasks	`GPT-6 LunaClaude Haiku 5.5DeepSeek V4.1 Flash`	low
deep-work	Long tasks at frontier depth	`Claude Opus 5.5GPT-6.1 Sol`	max
google-search	Research grounded in Google Search	`Gemini 4 ArgonGPT-6.1 SolClaude Opus 5.5`	medium
x-search	Live search on X	`Grok 4.7Kimi K3.1 ProGemini 4 Argon`	high
review	A different model tries to break the diff	`GPT-6.1 SolClaude Opus 5.5Kimi K3.1 Pro`	xhigh
swarm	Many parallel workers on separate slices	`DeepSeek V4.1 FlashGLM 5.3 FlashGPT-6 Luna`	low
forensics	Root-causing a live symptom or a captured trace	`GPT-6 AstraClaude Opus 5.5Kimi K3.1 Pro`	xhigh
synthesize	Merging many results into one answer	`Claude Opus 5.5GPT-6.1 SolGemini 4 Argon`	xhighxhighhigh

Every chain is yours to reorder. Edit ~/.omh/routing/model-chains.json, then run omh model-chains show to print what is in effect.

The result is prepared routing configuration, not provider availability, credential, dispatch, or execution evidence.

Read the routing setup guide

Flagship workflows

The ulw-* family.

Ten long-horizon lanes.

ulw-context Terminology alignment

Context

Aligns the words a repository uses before plans and handoffs.

Say ulw-context review project terms
ulw-interview Clarification

Deep Interview

One question at a time until the brief is clear.

Say deep-interview clarify
ulw-research Decision grounding

Research

Reference implementations, live web evidence, verified claims.

Say web research source-backed research
ulw-plan Reviewed plan

Ralplan

Consensus planning with review gates.

Say ralplan consensus plan
ulw-work Parallel delivery

Ultrawork

Splits an accepted plan into disjoint lanes.

Say ultrawork parallel work
dag-ulw Graph run

DAG ULW

Plans a graph of small nodes, runs them in parallel, verifies and repairs.

Say dag ulw /dag-ulw
ulw-maestro External handoff

Maestro

Runs the work on Claude Code or Codex, spawned live with a steerable session.

Say ulw-maestro coding handoff
ulw-loop Goal loop

Loop

Interview → plan → research → build → review.

Say loop long horizon goal
ulw-qa Adversarial QA

UltraQA

Hostile scenarios, end-to-end runs, release QA.

Say ultraqa release qa
ulw-perf Measured optimization

Ultraperf

Finds where the system is actually slow, leaking, or expensive.

Say ultraperf find the bottleneck

What you get

The right model for every lane.

`/dag-ulw` graph runs.

Expert skills on demand.

Long-term memory you can trust.

OMH Native.

Evidence before "done".

Just type `/dag-ulw`, and you're a graph engineering master.

Install once. Set up once.

Install the command

Set it up

Verify or troubleshoot

Be with us

Sponsors

Every lane on its own model

The ulw-* family.

Context

Deep Interview

Research

Ralplan

Ultrawork

DAG ULW

Maestro

Loop

UltraQA

Ultraperf

What you get

The right model for every lane.

/dag-ulw graph runs.

Expert skills on demand.

Long-term memory you can trust.

OMH Native.

Evidence before "done".

Just type /dag-ulw, and you're a graph engineering master.

Install once. Set up once.

Install the command

Set it up

Verify or troubleshoot

Be with us

Sponsors

Every lane on its own model

The ulw-* family.

Context

Deep Interview

Research

Ralplan

Ultrawork

DAG ULW

Maestro

Loop

UltraQA

Ultraperf

`/dag-ulw` graph runs.

Just type `/dag-ulw`, and you're a graph engineering master.