Hermes Pixel Office Features: Full Breakdown

The Hermes Pixel Office is a free, open source dashboard that draws every Hermes agent as a pixel character in a shared office, and it has more in it than the screenshots suggest.

This is the feature-by-feature breakdown: what each part does, what it is telling you, and how to use it.

It is built by Teknium, one of the people behind Nous Research — the team behind Hermes Agent itself. Nothing to sign up for, nothing to unlock.

What it is, in one line

A live window into every Hermes agent running on your machine, drawn as a cartoon office where each agent is a character at a desk.

They walk in, sit down, work, and walk out when the job is done.

It exists because Hermes agents normally run in the background — running tools, spinning up helper agents, doing real work — with nothing visible except scrolling text.

Feature 1 — One character per agent session

Every agent session is one character.

It does not matter where the session came from: the command line, a chat app, or a job you set to run on a schedule.

They all show up as their own little person walking in through the door.

That is the base unit of the whole interface, and it is why the room stays readable even with several jobs running.

Feature 2 — Gold collars and name tags for helper agents

When a bigger agent spins up a helper to handle a side task, that helper appears with a gold collar and a name tag showing its job.

So the second you look at the office you can separate the main workers from the sub-agents they pulled in.

For anyone running batch automations, this is the feature that makes the display useful rather than decorative — sub-agent activity is normally buried inside the parent's output.

Feature 3 — Live action animations

Characters act out what they are doing, so you can read state without touching a log.

Animation	What the agent is doing
Typing	Writing a file
Little book open	Reading something
Browsing	Using the web
Green monitor flicker	Running something in the terminal
Red exclamation mark	Waiting on your approval
Standing and leaving	Task complete

Feature 4 — Approval flags

Sometimes an agent hits a command that could be risky and needs a yes or no from you before it goes ahead.

That agent throws up a bright red exclamation mark over its head.

This is the feature that turns a silent blocker into a visible alert, and for automation it is the most valuable thing in the whole tool.

Feature 5 — The waiting counter

The top of the screen shows a counter of how many agents are waiting — something like "2 waiting".

So you do not need to scan the room to work out whether anything is blocked.

One number, top of screen, always current.

Feature 6 — The optional chime

There is a chime you can switch on that dings when an agent needs your okay, or when a helper finishes its job.

It is off by default.

Flip it on and you do not even have to be looking at the screen — which is the difference between supervising agents and babysitting them.

Feature 7 — One shared office

Every Hermes agent running on your machine shares the same office.

If you have several things going at once, they all pile into one room together.

One window, your whole team, no per-job dashboards.

Feature 8 — Browser page or VS Code panel

Two ways to run it, same office either way.

Browser: open it as its own page.

VS Code: drop it in and the office sits in a panel next to your code, plus a "+ agent" button that starts a fresh Hermes run without leaving the editor.

Feature 9 — Demo mode

It ships with a demo mode that fills the office with pretend agents.

Use it before connecting anything real, so you learn what every character and flag means without guessing on a live job.

Feature 10 — Remote support

It works over a remote connection out of the box.

If your agent OS runs on a server, as mine does, you still get the office on your local screen with no extra setup.

🔥 Want the Agent OS this plugs into?

Inside the AI Profit Boardroom you get the Agent OS zip ready to install with Hermes as the brain, a 30-day roadmap, and four live coaching calls a week with 4,000+ members.

→ Get access here

What it deliberately does not do

It does not slow your agents down. The Pixel Office is a window and nothing more — it watches what your agents are already doing and draws it. It never blocks them, never changes their work, and never adds weight to how they run. Your agents do not even know it is there.

It does not control anything. Keep your agents' own safety rules turned on. The red flag only tells you when to look; you still say yes or no to anything risky.

It is not an audit trail. It is a live read on state, not a record you can go back through. Keep your logging.

Using the features together on a real job

Batch content generation is where all of this lines up.

I ran a group of Hermes agents drafting hooks, scripts and captions — each helper appeared gold-collared, several typing and reading at once, and the red flag plus counter told me the exact second one needed a decision.

The same setup works for an onboarding sequence, where characters finishing and walking out confirms the whole flow completed, and for a lead system, where one glance shows who is working, who finished and who is blocked on me.

That is the whole point: supervision in a second instead of a log dive.

FAQ

What features does the Hermes Pixel Office have? One character per agent session, gold collars and name tags for helper agents, live action animations, approval flags, a waiting counter, an optional chime, a shared office for all agents, browser and VS Code modes, demo mode, and remote support.

How do I spot a blocked agent? A red exclamation mark appears over that character, and the counter at the top shows how many are waiting.

What do gold collars mean? They mark helper agents spun up by a bigger agent, with a name tag showing the job.

Does it add overhead? No — it only draws what agents already do, and never blocks or modifies their work.

Can I try it before connecting real agents? Yes, demo mode fills the office with pretend agents.

About Julian

I'm Julian Goldie — AI entrepreneur, SEO expert, and founder of the AI Profit Boardroom (4,000+ members). I help business owners scale with AI agents, automation, and SEO.