Quickstart
Your first mission
This walks the same path the app's first-run wizard does: open MUON, pick a workspace, connect one coding agent, run a real mission, watch the crew work, and decide the gates it files. Nothing here is destructive — MUON's seeded first task only adds files.
Run the loop
- 01
Open the app and pick a workspace
Launch MUON (see Install if you haven't yet) and choose the folder you want the crew to work in. Memory, blast-radius and every path in the run are scoped to it.
- 02
Connect a vendor
The first-run wizard checks Claude Code, Codex, Cursor and OpenCode separately for installed vs. authenticated, and gives one concrete next action for whichever is missing. You never hand MUON a token — you log in to the vendor's own CLI, and MUON only observes readiness.
muon onboard
- 03
Run a mission chat
Type what you want done into the chat. MUON's orchestrator plans the work, dispatches it to the fleet, and reports back — the gates stay yours. No task in mind? MUON can seed a tiny, safe sample so you can watch the whole loop.
muon quickstart # seed + dispatch a safe sample task muon chat --workspace <dir> # or start a real mission yourself
- 04
Watch the crew
Open the desktop's Crew Topology tab (⌘K → “Open crew topology”) to see who is doing what: one node per vendor lane, subagents nested under their parent, and peer edges only where two jobs actually exchanged an addressed message.
muon crew roles --chat <id> # who holds which role, and why muon crew coord --chat <id> # peer messages, file claims, conflicts muon dispatch status # active jobs across the fleet
Peer messages render as explicit “Agent text · untrusted” — another agent's words are evidence, never an instruction, and never something that can approve or dispatch on its own.
- 05
See the pre-edit view
Before an agent edits a target, MUON fuses its code blast-radius with the human-confirmed memory anchored to it — prior decisions, contested proposals, and live cross-agent activity on the same target.
muon context muon-hello.ts # or any symbol / file / module path
Trusted decisions show their full text. Contested proposals show existence-only until you explicitly ask to read them — an unvetted note can never reach an agent as if it were settled fact.
- 06
Approve, review, merge
When an agent needs to cross a boundary — an edit outside the expected surface, a network call, a merge — MUON files a gate and waits. A merge gate is graph-certified when possible: MUON reviews the diff itself and either certifies it or marks specific files REVIEW BLIND for a human read.
muon approve list muon approve review --approval-id <id> # artifact digest + blind files muon approve resolve --approval-id <id> --status approved
Honest about full-auto
The desktop sidebar's full-auto panel turns red when armed — that color is deliberate: it is disabling a gate, and MUON wants that visible at a glance, not buried in settings.
- Off (default): every approval waits for you.
- Armed, all lanes: approvals resolve automatically. Egress and blocked merge reviews still ask.
- Armed, a subset of lanes: checked lanes approve automatically; everything else still asks. A request MUON cannot attribute to a lane is never covered by a partial selection.
Standing consent, not a bypass
Every auto-approval goes through the same approval path a manual click uses and leaves a named receipt in the audit trail — the exact command or scope approved, redacted of secrets, with its risk level. See Governance for what full-auto can never cover.