Skip to content

armgit / mso

Public Imported from GitHub
0

Browser-based graphical shell and control plane for a Linux server you own — real terminal, file manager, live metrics and BYOK AI in one mobile-first workspace. Next.js 16 · React 19 · node-pty · Tailwind v4 · shadcn/ui. Public alpha, self-hosted.

4.5 KB
Raw Blame

First 10 minutes with MSO

This is the canonical tutorial: one path to first value. It deliberately does not explain every feature. For production hosting, use a supported Linux host and keep the raw app loopback-bound behind your chosen protected HTTPS/private-network path.

Goal

Reach one verified, read-first agent action against a real project, then see the evidence you can resume later.

1. Install and diagnose

curl -fsSL https://raw.githubusercontent.com/rahmanef63/mso/main/scripts/install.sh | bash
mso doctor

Do not continue past an unexpected doctor failure. Fix the named layer first.

2. Open MSO

mso web

Approve the device through the supported local/Owner flow if this browser is new. Treat Owner access like SSH-equivalent authority for the MSO service user.

The Set up MSO guide opens for a newly signed-in browser. An existing browser that already closed onboarding keeps that choice. Reopen it at Settings → Account → About → Maintenance → Setup. The guide retains its step, selected project and run reference in this browser; skipping it does not reset the host or require a new account or team. Viewer/Operator devices see access readiness without the Owner-only model and project-check controls.

If restoring existing data, choose Open backup options before creating work. The current Settings surface can import browser data and verify server memory snapshots into an isolated folder. Neither operation is a full installation restore; use the supported installation and recovery procedures for that scope.

3. Connect one model/provider

In the guide's Readiness step, review the service user, filesystem roots and access route. Open model connections opens the existing private Integrations form in another tab. Configure one provider/model, return to the guide, and refresh. The explicit Test model request action uses the saved selection and may consume provider quota. Stored connection metadata alone is not proof of inference, tool calls or multimodal support. Model setup can be skipped for the first project check. Do not paste provider credentials into agent chat.

4. Pick one project

Use the Projects/Files surface or the CLI to identify a real project. Start read-only: inspect its status, current branch and recent work before asking an agent to edit anything.

The guide's Project step lists a bounded page from the existing host discovery service and supports searching by project name. Empty, failed and truncated discovery are shown separately. Selecting a returned project retains its exact id; execution resolves it again through the host guards.

Choose Review first action → Run project check to save a manual workflow and read that project's Git, package and knowledge metadata. The run uses the existing Workflow Graph engine and project_get, with no project-file mutation or model call. The guide displays completion only when the persisted receipt confirms all expected steps and the selected project. A lost response retains the same request identity; reopening a known receipt reads it without dispatching again. The saved workflow remains available under Workflows. This preflight is separate from the agent task below.

5. Run the first safe agent task

From an MCP client or MSO agent, use a request shaped like:

Check this project's current status and explain the highest-value next task.
Do not change files, branches, services, deployments, credentials or external systems.

The first success criterion is not “the model answered.” It is that MSO resolved the project, used bounded reads, and produced inspectable execution evidence without widening authority.

6. Make one isolated change

For source-changing work, use one task-specific hidden worktree. Never use canonical main as a shared agent scratchpad. Inspect the diff and run the relevant tests before integration.

7. Verify and resume

Confirm the session/workflow evidence exists and can be resumed. A completed response is not equivalent to a tested, integrated, released or live-verified change.

Where next