A 3D isometric office where AI agents do real work on your own Claude login.
Six departments, thirty-five agents at their desks, a task bar that routes what you type to the right agent, and a Brain at the centre that is your own folder of notes. Type a task, the office gives it to the right person, they read your notes, use the connectors you have already set up in Claude Code, do the work, and file the result back into your notes. Everything runs on your machine.
Beta. It works end to end. Expect rough edges and tell us about them in Issues.
License, in plain English: free for personal and internal use. You may not sell it, resell it, or build a paid product on it. (Formal terms: PolyForm Noncommercial 1.0.0 — see LICENSE.)
- macOS or Linux (Windows: works with
npmcommands directly,./setupis Bash only) - Node.js 20+ — https://nodejs.org
- git
- Claude Code, logged in with your Claude account, or an
ANTHROPIC_API_KEY
git clone https://github.com/ajsahni/agents-office.git
cd agents-office
./setup # checks Node, git and Claude; installs; builds; boots once
npm start # → http://localhost:4520Without ./setup: npm install && node build.mjs && npm start.
- Open http://localhost:4520. The panel on the right says LIVE · CLAUDE when the server is
connected. Double-clicking
dist/command-centre-v2.htmlopens the same office on its own, without a server, in demo mode. - In the bar at the top of the panel, pick a department, type a task in plain words, press Add. Claude picks the agent and names them; the task appears in the feed; the agent picks it up, works, and the deliverable lands in that agent's chat and in your brain folder as a note.
- Click any agent to talk to them. They answer in their role, grounded in your notes.
Say
revise: make it shorterand they rework their last deliverable. - Press G, or click the Brain, to open your notes as a graph. Hover a note to see its links, click it to read where it sits and who read or wrote it.
- The top bar shows the connectors your Claude Code is connected to. When an agent uses one, its logo pulses and the wire into that department lights up.
- The first deliverables will be competent and generic: the agents know the sample studio and one sentence about their own job. Point the brain at your notes, then teach them how you work (below). That is where the office becomes yours.
The bar under CONNECTED TO is real: it is the list from claude mcp list on this machine,
which is the same list the agents get as tools. Gmail, Slack, Notion, Google Drive, Canva,
whatever you have connected in claude.ai or added with claude mcp add. A server that needs
authentication shows grey with the reason on hover, and is not wired to any pod until it works.
Nothing connected yet? The bar says so.
Agents can call those servers while they work, plus web search. They never get Bash, file tools or sub-agents. Their standing rule: read freely; send, post, pay, delete or change anything outside this machine only when your task explicitly asks for that exact action. A finished deliverable says which tools it used, and the note in your brain records them.
Decide what the agents may touch in office.config.json:
"mcp": { "allow": [], "deny": ["Stripe"], "departments": { "Slack": ["emails", "ops"] } },
"tools": { "web": true }allow empty means every connected server. deny keeps a server in the bar but out of the
agents' hands. departments says which pods a server is wired to (known brands have a default;
anything else feeds every pod). Set tools.web to false to keep the agents off the web.
Tool use needs the Claude Code login; on an ANTHROPIC_API_KEY the agents write from your notes only.
The 35 agents are in office.agents.json: an id, a department, a name, a role, what they do,
and the connectors they usually use. Change the name, the role, what they do and their tools.
Departments, leads and seats are fixed: six pods, 35 desks, that is the office. A new kind of
agent is a renamed seat in the right department.
The easy way is to let Claude do it. Open Claude Code in this folder and say what you want:
claude
> Rename the Newsletter agent to PODCAST NOTES. It turns each episode into show notes and a LinkedIn post, and uses Google Drive.
> Make the Sales department about wholesale accounts, not inbound leads. Rewrite what each agent does.
> Tell every Finance agent to use Xero and nothing else.
Claude reads CLAUDE.md, writes your changes to office.agents.local.json (yours, ignored by
git, so git pull never overwrites it), and validates them with npm run check. Restart the
office and the desks carry the new names. Edit the file by hand if you prefer; the shape is:
{ "agents": [
{ "id": "newt", "name": "PODCAST NOTES", "role": "Podcast Notes Agent",
"does": "Turns each episode into show notes and a LinkedIn post.", "tools": ["google drive"],
"brief": "Show notes are five bullets and a pull quote. The LinkedIn post opens with the quote, never with the episode title." }
] }Edits to id, department or lead are ignored, and the server says so at start. The same
file can also sit in your brain as <brain>/Agents Office/agents.json; the office reads the
shipped roster, then the brain's, then the local file.
Renaming an agent says what it does. It does not say how. Out of the box every agent knows its one-line job, your notes, and a rule to hand over a finished deliverable, so the first results are competent and generic. Two ways to fix that, both read before every task:
- A brief is a few standing sentences on one agent: tone, red lines, who to escalate to.
It is the
brieffield above. - A skill is a folder in your brain,
<brain>/Agents Office/skills/<name>/, with aSKILL.md(when it applies, the steps, the shape, the rules) and the template or example beside it. Bind it to an agent or a department in its front matter. Same shape as a Claude Code skill.
---
name: proposal
description: How we write a client proposal
agents: [piper]
---
# Writing a proposal
Use this for any request that ends in a document a client says yes or no to.
1. Prices come from `10-Business/offer-ladder.md`. Never invent one.
2. Follow `template.md` beside this file, section for section.
- Three options, always. Recommend the middle one.Three example skills ship in skills/ for the sample studio. The fastest way to write yours is
to hand Claude Code what you already have, the SOP, the email you keep copying, the last report
you were happy with, and ask for a skill:
claude
> Here is the proposal I sent Harbourside. Turn it into a skill for the Proposals agent: the
shape as a template, the rules I follow, and keep this one as the example.
Skills and briefs take effect on the next task, no restart. npm run check validates them and
http://localhost:4520/api/skills shows who has what. The full guide, including what the agent
sees and how to write a good one, is SKILLS.md.
Click a department lead and say set up. The lead asks five questions, one at a time: what the department does here, the job you do most, what a good result looks like, what must never happen, which tools and people are involved. Then it writes a brief for each agent on its team and a skill for the job you described, into your brain, and tells you exactly what it wrote and one task to type to try it. Nothing is written until the last answer. "skip", "done" and "cancel" do what they say. A lead whose department has nothing of yours yet offers this in its greeting.
Send a deliverable back with revise: … in the agent's chat and the correction is recorded in
<brain>/Agents Office/feedback/<agent>.md. Claude sorts it: a one-off about that task, or a
standing rule ("proposals are always one page") that the agent then applies to every task from
then on. The file is plain Markdown and it is yours: reword a rule, delete a line to unlearn it,
move a one-off up to make it a rule. When a rule is really a process, ask Claude Code to fold it
into the skill.
office.config.json:
{ "name": "Northgate Studio", "brain": "./brain", "port": 4520, "model": "" }- name — your business. It appears in the title and in every agent's brief.
- brain — a folder of Markdown notes with
[[wiki links]]. An Obsidian vault works as is. The sample brain inbrain/is a small fictional studio so the office works out of the box. Point this at your own notes and rebuild (node build.mjs) or just restart the server. - port — where the office listens.
- model — leave empty for your Claude Code default, or name a model.
Put private overrides in office.config.local.json (ignored by git).
Agents write their deliverables to <brain>/Agents Office/ as dated notes with a link back to
every note they read, so your graph grows as the office works.
| Key | Does |
|---|---|
1 to 6 |
Marketing, Emails, Sales, Operations, Finance, Delivery |
B |
The company board: every department, backlog to done |
G |
The Brain graph |
C |
Chat with the department lead |
X |
Send two agents to meet at the Brain |
V |
Full screen view with dimmed lighting |
D |
Dark mode. http://localhost:4520/dark opens in it |
Esc |
Back |
npm run check # build, offline smoke test in a headless browser, server smoke test
npm run check:live # the same, plus one real task and one chat turn through ClaudeEvery check prints ✓ or ✗ with the reason. The Beta was built against this loop and it is the first thing to run after any change.
| Path | What |
|---|---|
src/ |
The office: main.js scene, tasks.js task panel, brain.js the Brain, mcp.js connectors, data.js departments and roster, v1data.js agent personalities |
serve.mjs |
The local server: routing, deliverables, chat, the live Brain graph |
mcp.mjs |
Connectors: claude mcp list parsed, allow/deny, the tools each agent may call |
roster.mjs · office.agents.json |
The 35 agents: names, roles, what they do, their tools, their briefs (<brain>/Agents Office/agents.json and office.agents.local.json override) |
skills.mjs · skills/ |
Skills: how a kind of work is done, bound to agents or departments (<brain>/Agents Office/skills/ is yours) |
learn.mjs |
Corrections from revise: … recorded per agent in <brain>/Agents Office/feedback/; standing rules go back into the prompt |
onboard.mjs |
The lead's five-question set-up interview; writes briefs and a skill into the brain |
SKILLS.md |
The guide to briefs and skills |
CLAUDE.md |
What Claude Code does when you ask it to change agents, write a skill, or change connectors in this folder |
graph-build.mjs |
Reads your brain folder and lays out the graph |
dist/command-centre-v2.html |
The office as one built file (node build.mjs from src/); the server serves it, or double-click it for the demo |
brain/ |
The sample brain |
data/tasks.json |
Your tasks (created on first run, ignored by git) |
Your notes are read from disk and sent to Claude only as context for the task or chat at hand
(a handful of the most relevant notes, plus your brain's CLAUDE.md and index.md if present,
plus the agent's brief, skills and standing rules from your corrections).
When an agent calls a connector, that call goes to that service through your own Claude Code
login, exactly as it would if you called it yourself. Nothing else leaves your machine.
Deliverables are saved locally.
