Product · Station

The desktop runner. Native to your machine.

Station spawns Claude Code agents on your workstation, dispatches tasks from the queue, and streams output back to the dashboard. Architecture scanning runs locally, so your source never leaves the machine. Native macOS and Windows.
Meshly Build · Station
v1.2.7
HOME
AGENTS
PROJECTS
LOGS
SETTINGS
CONNECTED · server.meshly.build
Active agents · 3
⌘K
◆
Orion
auth refresh
RUNNING
◆
Atlas
orders test suite
RUNNING
◆
Nova
idle
IDLE
01 · Tauri-based desktop

A desktop app, not an Electron tab.

Station ships as a Tauri app: a Rust binary with a system-webview UI, not a bundled Chromium. Universal macOS build (Intel + Apple Silicon) plus .msi and .exe installers for Windows. Tray icon, menu controls, and a real native presence. Smaller, lighter, and lower-overhead than the Electron alternative.

  • Tauri v2. Rust backend, native webview frontend. No bundled Chromium.
  • macOS universal binary; .msi and .exe (NSIS) installers for Windows.
  • Tray icon with Show / Hide, Pause All, Resume All, Refresh MCP, Settings, Quit.
macOS
13+
Universal binaryTray menuTauri.dmg installer
Windows
10 / 11
.msi + .exe (NSIS)Tray appTauriHeadless mode
02 · Architecture scanner

Scan locally. No extra setup.

The architecture scanner runs in-process inside Station. No separate CLI to install, no Python toolchain. The scan walks your repo, detects surfaces, and builds a fixed-schema envelope that uploads to Server for the team. Your source never leaves the machine.

  • Built in to Station. Formerly a separate `pipx install meshly-build`, now native Rust.
  • Fixed envelope schema with `repo_sha`, `scan_id`, and timestamps for reproducible bookkeeping.
  • Regenerate on demand. Server stores the result so the team sees the same thing.
ARCHITECTURE SCAN · ledger-svc · illustrative
✓
Walk source tree
in-process
✓
Detect surfaces
rust pipeline
✓
Extract dependencies
rust pipeline
✓
Build envelope
fixed schema
↑
Upload to Server
authenticated
No external service. No extra install. Native to Station.
03 · Live streaming

Watch the agent work. From any browser.

Agent output streams from the workstation to the dashboard in real time. The full Claude Code interface (edits, test runs, decisions) visible to anyone with access. Streams are resilient across reconnects, sleep/wake, and deploy restarts.

  • Stdout, file edits, and tool calls stream live, not via polling.
  • Resilient transport survives laptop sleep, network blips, and server restarts.
  • Non-developers (PMs, engineering managers) can watch without an operator install.
Orion · streaming
→ DASHBOARD
● claude-code · session 7c1f
› Editing apps/api/src/auth/refresh.ts
↳ +28 −12
› Running pnpm test --filter @meshly/auth
↳ 84 passing · 0 failing
› Submitting for review
↳ task #4821 · status Review
maya · MacBook Pro · M3
⇄ server.meshly.build
04 · Command palette

⌘K everything.

The command palette opens on ⌘K and covers running tasks, switching projects, and jumping between views. A tabbed Settings page keeps general, project, and about configuration in one place.

  • Fuzzy search across tasks, projects, and recent actions, keyboard-first.
  • Tabbed Settings (General / Projects / About). Nothing buried two levels deep.
  • Sidebar nav: Home, Agents, Projects, Logs.
⌕run task⌘ K
◆
Run task #4830 with backend-dev
↵
◱
Switch project · ledger-svc
⌘ 2
◉
Open Logs
⌘ L
◈
Refresh MCP connection
05 · Headless reliability

Runs unattended. Doesn't hang on you.

Headless agents handle permission prompts without freezing, recover from sleep/wake cleanly, and clean up their own log files. Long-running CI usage works as well as interactive operator usage.

  • Headless agents auto-handle permission prompts. No operator-attended hangs.
  • Reconnects after sleep / wake or network blips happen automatically.
  • Log files clean themselves up; structured logs with on/off toggle and live level switching.
HEADLESS RELIABILITY · BUILT IN
Auto-handled permission prompts
Claude runs with `--permission-mode dontAsk` so headless sessions never hang on a TUI confirmation.
Reconnect with exponential backoff
WebSocket feed retries 2s → 60s after sleep / wake or network blips.
Auto-cleaned logs
Log files removed on successful task completion; old logs swept on Station startup.
Live log-level switching
Tracing subscriber updates the active filter from the UI without restarting Station.
Three pieces. One platform.

Station is the GUI runner. One of three.

STATION · DESKTOP
GUI runner
Tauri app on each operator's workstation. Spawns Claude Code agents, scans architecture, streams output. Native macOS and Windows.
CLI · AIR-GAPPED
Local-AI runner
Terminal-native runner that uses your own local model. For environments where cloud AI isn't an option. Read →
SERVER · MANAGED
Control plane
The shared workspace. Boards, agents, review, audit, knowledge, operated by Meshly.Read →

Put Station on every machine that ships code.

Station is the agent runtime; Server is the managed control plane. Tell us about your team and we'll walk through Station rollout with your real environment, including how it fits your managed-software setup.