By taskyou
Launch and manage AI agent teams on remote servers through an interactive setup assistant, then monitor task status, run diagnostics, and repair daemon issues from Claude
Check on all tracked tasks for an immediate status update
Check the health of your TaskYou-OS installation — plugin version, TaskYou version, daemon status, and executor health. Diagnoses and fixes issues automatically.
Diagnose and fix problems with the agent system
Quick reference of GM commands and how everything works
Getting started guide — what the GM can do and example prompts
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Your own AI agent team, running on a remote server and managed through Claude Code.
You describe what you need in plain English. A "General Manager" running in Claude Code delegates work to background agents on a remote server. They research, analyze, write, and build — even while your laptop is closed.
You need Claude Code. Open it and run:
/plugin marketplace add taskyou/taskyou-os
/plugin install taskyou-os
/taskyou-os:launch
That's it. The setup walks you through everything:
No git cloning, no manual config editing, no SSH knowledge required. The setup handles SSH keys, exe.dev signup, software installation, and authentication transfer for you.
You can re-run /taskyou-os:launch anytime to resume an interrupted setup or fix issues.
Your Machine Server
┌──────────────┐ ┌──────────────────┐
│ Claude Code │ ──── SSH ─────→ │ TaskYou daemon │
│ (GM session) │ │ ┌──────────────┐ │
│ │ │ │ Agent 1 │ │
│ │ │ │ Agent 2 │ │
│ │ │ │ ... │ │
│ │ │ └──────────────┘ │
│ │ │ │
│ │ │ notifications │
└──────────────┘ │ .jsonl │
└──────────────────┘
/gm-babysit for an immediate status check on all tracked tasksInstead of running the GM locally over SSH, you can deploy it entirely onto an exe.dev VM. This gives you a web-accessible GM with:
Browser exe.dev VM
┌──────────────┐ ┌──────────────────┐
│ Landing page │ ──── HTTPS ───→ │ nginx (port 8000)│
│ (kanban) │ ← board.json ── │ │
│ │ │ TaskYou daemon │
│ GM Terminal │ ──── xterm ───→ │ ┌──────────────┐ │
│ (browser) │ │ │ Agent 1 │ │
│ │ │ │ Agent 2 │ │
│ │ │ │ ... │ │
│ │ │ └──────────────┘ │
└──────────────┘ └──────────────────┘
To deploy:
# Set EXE_DEV_VM_NAME in your config.env, then:
./setup.sh exe ~/Projects/gms/myproject
This creates the VM, uploads everything, and prints the URL. Share access with teammates via ssh exe.dev share add <vm> [email protected].
Once your GM is running, these commands are available:
| Command | What it does |
|---|---|
/gm-babysit | Immediate status check on all tracked tasks |
/gm-status | Daemon, tasks, and agents overview |
/gm-fix | Diagnose and fix agent system problems |
/gm-start | Getting started guide with project-specific examples |
/gm-help | Quick reference for all commands and TaskYou CLI |
These run from the taskyou-os plugin context (not inside a GM):
| Command | What it does |
|---|---|
/taskyou-os:launch | Interactive wizard to create a new GM |
/taskyou-os:doctor | Health check — updates plugin, checks versions, daemon status, executor health, template drift, security audit |
Run /doctor periodically to keep things healthy. It also detects new commands available from the plugin and offers to add them to existing GMs.
The taskyou-os plugin updates through the Claude Code marketplace. Run:
/plugin marketplace add taskyou/taskyou-os
This pulls the latest version. Restart Claude Code afterward for changes to take effect.
npx claudepluginhub taskyou/taskyou-os --plugin taskyou-osMarketplace for the Agency CLI - an AI agent orchestrator
Multi-agent orchestration with AI SDK v5 - handoffs, routing, and coordination for any AI provider (OpenAI, Anthropic, Google)
Portable multi-agent harness for .agents-based skills and workflows
Multi-agent orchestration for Claude Code. 12 specialized agents working in parallel — planning, building, reviewing, debugging. Plus a Hub for always-alive multi-project sessions controllable from Telegram or Slack.
AI team role and worker manager for multi-agent development workflows.
Task distribution, agent coordination, progress monitoring - executes plans via subagents. Requires AI Maestro for inter-agent messaging.