The Muster field guide
Muster documentation
Muster brings your AI teammates, conversations, tools and approvals into one workspace. Start in the managed web app, connect your own provider API key, and add supported device and companion connections as your work needs them.
A little company for the first step.
- Open your workspaceStart with a place for your work.
- Connect a modelBring your provider connection.
- Give it a small taskRead the reply. Decide what comes next.
Your first five minutes
- Open Muster and follow the sign-in and setup steps.
- Connect your provider API key and select a model for your first bot.
- Send one small task, review the reply, and respond to any approval request.
Muster is a free app. Provider usage and connected services can charge separately. Follow Quick start for the first conversation or Account & device setup for connection and storage details.
Start here
Start
Quick start
Open your workspace, connect a provider, and give your first bot a task.
Start
Install & apps
Desktop downloads for macOS, Windows and Linux, auto-updates, and the mobile companions.
Start
Account & device setup
Connect provider keys and devices, review Drive consent, and understand workspace storage.
Operate
Engines & models
Connect OpenAI-compatible or Anthropic-compatible providers, pick models per bot, watch health.
Operate
Goals & autonomy
Goal mode: give a bot an outcome and let it loop — with hard stops, budgets, and provenance chips.
Operate
Automation
Routines, sentries, routine scorecards and the why-journal: work that keeps happening without you.
Operate
Approvals & privacy
Approval cards, standing grants, the evidence strip, and how workspace data is handled.
Reference
For agents
The skill file and llms.txt that let coding agents and AI tools operate Muster for you.
Reference
Security & license
Data handling, control limitations, issue reporting and applicable licenses.
What's new
- Goal mode — set an outcome instead of managing turns. Bots loop plan → act → verify with a turn budget, pause/stop controls, and a banner that always shows what the goal is and what it's costing you.
- Provenance chips — every bot reply is stamped with the engine instance and model that produced it. Hover the chip to see exactly who said what.
- Routine scorecards & why-journal — recurring bots keep a HYPOTHESIS/FINDINGS journal and get scored on their routines, so you can see drift before it becomes a problem.
See Goals & autonomy for the full picture, or grab the latest build on the download page.
Adding a device?
Use the account and device guide to review your execution target and permissions. Existing desktop data, hosted workspace data and backups have separate boundaries; signing in is not a complete migration or synchronization step.