On Air · Your Company Is Live

Run an AI company.
Not a prompt.

Open-source, local-first, MIT. Hire from 57 hand-written role specs , build a real org chart with hierarchy, file tickets, schedule work, and watch agents operate in live token streams. Your data never leaves your machine. Zero phone-home, ever.

Download v3.4.0 View source Quick start Changelog

57 ROLES

10 PROVIDERS

MIT LICENSE

LOCAL-FIRST

V3.4.0

AtlasGo

Engineering Lead

TOK 14,208

$0.00

VegaExec

Research

TOK 9,640

$0.00

OrionHold

Operations

TOK 2,110

$0.00

NovaGo

Data

TOK 5,902

$0.00

GGUF llama-3.1VRAM 11.4 GBTOK/MIN 1,840● LOCAL-FIRST

MOD · THE ORG · 01 S/N · TX-ROLES-0057

Hire from a curated F10 role library

Six levels of hierarchy. Not generic templates.

Fifty-seven hand-written role specs across Officer, Senior Management, Management, Supervisor, Lead, and IC. Every spec is its own artifact: preferred model tier, allowed and denied tools, cooperation patterns, decision authority. Drop one into your org, set their manager, and they slot in like a person joining a team. Not like a prompt being added to a queue.

Sample role spec Ready

--- senior-fullstack-engineer.md ---
role_id: senior-fullstack-engineer
name: Sarah Chen
level: Senior Management
preferred_model_tier: claude-opus
tools_allowed:
  - file.read
  - file.write
  - git.commit
tools_denied:
  - network.request
capabilities:
  - decompose_project
  - delegate_subtask
  - review_deliverable
---

You are a Senior Full-Stack Engineer. You write
production-grade code, mentor mid-level engineers,
and own delivery of features end-to-end…

ROLES

LEVELS

PROVIDERS

TESTS

3,521

MOD · LIVE COCKPIT · 02 S/N · TX-COCKPIT-0001

Watch the team operate. Live.

Every signal is data-bound. Nothing is decoration.

Token streams from every active turn. Tickets sliding from Open to In Progress as the orchestrator dispatches them. A ReAct-style agentic loop answering grounded to your data with hard step, token, and wall-clock budgets. Five dashboard subviews. One Cmd+K palette. A real cockpit, not a chat wrapper. And as of v3.4.0: arm Proactive Mode from the dashboard and idle employees find their own next ticket.

Org Chart

Build a real organization. Levels, reports, role families. Hire from 57 hand-written specs, not generic templates.

Live Streams

Exec

Watch every agent work in real token streams. Plan, tool call, result, answer. No black box.

Tickets & Meetings

Hold

File work, schedule it, call all-hands. The orchestrator dispatches; you command.

AGENTS

07/ 08 LIVE

1 idle: Tyler Hu

TICKETS

24/ 30 OPEN

3 in flight, 2 blocked

MTGS/H

4.2PER HR

avg 18 min, 2 in queue

COST/H

$0.00LOCAL

Ollama local, zero cloud

OpenHold

TKT-218

Wire Cmd+Shift+K to copilot ask flow

RV · P2

TKT-219

Add tabular-nums to telemetry totals

unassigned · P3

TKT-221

Provider router cost rounding for Groq

unassigned · P2

In ProgressExec

TKT-214

Decompose Q3 launch project into milestones

SC · P1

TKT-216

Migrate FTS5 index to canonical name

DE · P2

BlockedWarn

TKT-211

Pinned Ollama version on macOS arm64

DE · waiting on upstream

DoneGo

TKT-208

Copilot weights changed audit event

SC · closed

TKT-205

Ticker pause-aware on meeting.started

DE · closed

TKT-204

RAG attribution dedupe on SHA256

DE · closed

FLIGHT TKT-218AT OPENQUEUE 09● SYNC OK

PlanWhy is the frontend team behind schedule? Need frontend tickets in In Progress longer than 7 days, plus blockers and assigned employees.

Toolquery_tickets(team="frontend", status="in_progress", min_age_days=7)

Result3 matches · TKT-214 (12d, SC), TKT-216 (8d, DE), TKT-217 (9d, DE)

Toolquery_employees(level="lead", department="frontend")

Result2 matches · Sarah Chen, Dev Engineer · combined load 0.91

AnswerThe frontend team is at 91% load with 3 tickets aging past a week. SC is bottlenecked on the Q3 decomposition (TKT-214, 12 days open). Recommend either escalating TKT-214 to a stakeholder review meeting or temporarily redirecting DE off TKT-216 to unblock it.

MODEL llama-3.1BUDGETS STEP · TOK · WALL● GROUNDED

Ask anything. A ReAct-style loop with hard step, token, and wall-clock budgets answers grounded to your data: plan, tool, result, answer. No prompt theater, no black box.

Download Team-X

Team-X dashboard in Night Ops: Mission Control panel with agent runs, employee queues, and six KPI wells showing live runs, external runtimes, workforce active, queue pressure, blocked work, and today's cost — The shipped build, not a mockup. Mission Control in Night Ops; the full walkthrough with more frames is on the product page.

You don't manage prompts. You run a company. - Team-X README · v3.4.0

MOD · PRIVACY POSTURE · 03 S/N · TX-PRIV-0000

Local-first by default. Zero phone-home.

Your machine is the perimeter.

Your data lives in a SQLite database and a filesystem vault on your machine. Secrets sit in the OS keychain via keytar, never in plaintext config files. Choose which provider tiers your agents are allowed to use. Run the entire org on Ollama with no cloud dependency at all. Updates are user-triggered. No analytics. No telemetry. No crash reports. No exceptions.

OllamaLocal

Local · privacy-pure

llama3.1:8b · phi3 · mixtral · qwen2.5

LM StudioLocal

Local · privacy-pure

any GGUF · MLX

GroqOSS

Open-source cloud

llama-3.1-70b · mixtral-8x7b

OpenRouterOSS

Open-source cloud

mixtral · llama-3 · deepseek

TogetherOSS

Open-source cloud

llama-3.1-405b · qwen2.5-72b

FireworksOSS

Open-source cloud

deepseek · qwen · llama-3

OpenAI-compatOSS

Bring your own

any compatible endpoint

AnthropicCloud

Proprietary cloud

claude-opus · claude-sonnet · claude-haiku

OpenAICloud

Proprietary cloud

gpt-5 · gpt-4o · gpt-4o-mini

GoogleCloud

Proprietary cloud

gemini-2.5-pro · gemini-flash

PHONE-HOME

ANALYTICS

CRASH RPT

TRACKERS

MOD · INSTALL · 04 S/N · TX-DEPLOY-0004

Three platforms. No account.

Native installers, or build from source.

Native installers for Windows, macOS, and Linux. Or build from source with pnpm. No sign-up wall. No credit card on file. Run on Ollama with zero cloud, full feature parity, fully offline.

MOD · Install 03 PLATFORMS

Windows

NSIS installer

x64 826 MB arm64 830 MB

macOS

DMG · right-click, Open on first launch

Apple Silicon 124 MB Intel 130 MB

Linux

AppImage needs FUSE 2 · .deb does not

AppImage x86_64 274 MB .deb amd64 164 MB

RELEASE v3.4.0 LICENSE MIT INSTALL DOCS ● VERIFIED

Or run locally via Ollama

# Prereqs: Node 22+, pnpm 9+
git clone https://github.com/git-rocky-stack/team-x.git
cd team-x
pnpm install
pnpm dev

# Local model
ollama pull llama3.1:8b
ollama serve

§ 05 · Field Notes

ENGAGE · FINAL HANDOFF · CONSOLE READY S/N · TX-ENGAGE-v3.4.0

§ Mission · Install & Engage