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.
MSO documentation map
This index separates current reference from historical evidence so a dated plan cannot quietly become today's architecture documentation.
Authority order
- Current code and live descriptors (
/api/health,GET /mcp). - Generated contracts (
CLI.md,CHANGELOG.md). - Current reference docs below.
PROGRESS.mdfor why/when a change shipped.- Historical plans/audits for their dated context only.
Run node scripts/check-docs.mjs to validate relative links plus selected
machine-verifiable facts, and node scripts/gen-comparison.mjs --check to validate comparison
evidence and review freshness. The repository quality gate runs both automatically.
Current reference
| Document | Contract |
|---|---|
../README.md |
Product overview and quickstart |
QUICKSTART.md |
Canonical first-10-minutes tutorial: install → doctor → provider → project → safe first agent action → evidence |
LAUNCH-READINESS.md |
20 measurable product/launch gates, P0/P1 policy and truthful launch-label contract |
LAUNCH-QA.md |
Concise launch answers for category, competitors, agent concepts, security, memory, recovery and readiness |
SOURCE-CONCURRENCY.md |
One-task-one-worktree, canonical integration authority, remote protection and concurrency acceptance tests |
AGENT-SECURITY.md |
Agent-specific threat model, tool risk classes, abuse cases and memory-admission security contract |
DISASTER-RECOVERY.md |
3-2-1 target, RPO/RTO objectives, backup classes and restore-drill acceptance |
RELIABILITY.md |
Initial SLIs/SLO targets, error-budget policy, disk thresholds and release reliability contract |
ARCHITECTURE.md |
Current runtime, shell, role, host, Agent/Local Agent/subagent, MCP and deployment architecture |
GATEWAY.md |
Provider-neutral gateway state, health, ownership, supervisor and deployment contract |
COGNITIVE-RUNTIME.md |
Provider-neutral MCP/session/context/tool-routing runtime and benchmark contract |
AGENT-BOOTSTRAP.md |
Canonical first-call sequence and capability-parity matrix across CLI, MCP, and Alfa |
AGENT-CODING-PLAN.md |
Audited coding-capability gaps in the MCP catalog, the four tools added to close them, and what stays out of scope |
AGENT-PTY-DESIGN.md |
Why an agent-driven PTY is the highest-blast-radius tool in the catalog, and the bounds that keep it equal to exec_run |
AGENT-EDIT-PREVIEW.md |
Why fs_edit returns the diff it applied, what a dry run promises, and how the TUI renders it |
RASMIC.md |
Risk-aware orchestration, repo-local memory/evidence, collision detection and recipe/script promotion |
MEMORY-GRAPH.md |
Owner Memory app: vault notes, project knowledge, agent memory, layouts and limits |
AGENT-VAULT.md |
Separate private repository agent-progress vaults, source/data boundaries and Obsidian use |
MEMORY-SAFETY.md |
Session-source preservation, owner cleanup previews, local memory backups, learning/discovery visibility and remaining limits |
COMPARISON.md |
Generated product comparison methodology, evidence and notes |
references/README.md |
One file per upstream project MSO studies or pins: license, pin, what was used and what was not copied |
LICENSE-HISTORY.md |
Current PolyForm Noncommercial license, the permanent prior MIT grant, and open contribution questions |
COMPETITIVE-ROADMAP.md |
Executed comparison plan, deliberate specialist boundaries and next investments |
INSTALL.md |
Owner installation, TLS, updates, backup and uninstall |
PLATFORMS.md |
Linux/macOS/Windows/Android/iOS support modes and machine adapters |
TERMUX.md |
Android/Termux bootstrap through Ubuntu PRoot, repair path and verification |
INFRASTRUCTURE-PROVIDERS.md |
Interactive MSO Agent TUI/approval/error contract plus Dokploy/Cloudflare/Hostinger/Composio credentials, tools and deployment boundary |
DEVELOPMENT.md |
Development and release workflow |
SECURITY-ASSURANCE.md |
Repeatable security verification lanes, public evidence, and claim boundaries |
CLI.md |
Generated mso CLI contract |
INTEGRATION-PORTABILITY.md |
Reviewed JSON identity transfers, encrypted credentials, SC interoperability and receiver limits |
INTEGRATIONS.md |
Native temporary credential forms, direct HTTPS secret submission, CLI and ChatGPT Page |
INTEGRATIONS-NATIVE-GOOGLE.md |
Native Google OAuth app/account separation, Search Console and GA4 reads, browser binding, scope and parity limits |
CHANNELS.md |
Provider-neutral Telegram/Discord messaging, inbound verification, outbound send and Workflow bridge |
PROJECT-MCP-CONNECTIONS.md |
Named private connections, endpoint binding and identity for modular project MCPs |
BATONLY-FEDERATION.md |
Pull-based Batonly capability federation worker, MSO/SI-Coder execution boundary, polling, security and acceptance contract |
AUTOMATION-FLOWS.md |
Versioned project API/MCP flows, CLI, sessions, Add MCP, assets and limits |
WORKFLOW-SCHEDULES.md |
Interval, cron and durable one-shot scheduling semantics, missed ticks and recovery limits |
WORKFLOW-GRAPH.md |
Private server-native Workflow Graph v2 with n8n-core triggers/flow control, versions/history, variables, integrations, visual execution and automatic learning |
SHARED-GRAPH-FOUNDATION.md |
Canonical shared graph mechanics, feature adapters, ownership matrix and dense-graph Focus/Map interaction boundary |
NAKAMA-PARITY.md |
Frozen Nakama scope, current implementation mappings and candidate evidence limits |
N8N-PARITY.md |
Core n8n capability map across native Workflows, Organization, Integrations and the dedicated external n8n workspace |
TUI-PARITY-PLAN.md |
MSO Agent TUI parity program: per-item status, the server seam each item uses, and the items deliberately refused |
TUI-RENDERER-PLAN.md |
What the TUI is today, why a retained renderer is the only path to panels and tabs, and the M0 to M8 order that keeps every gate green |
TUI-HANDOVER.md |
Where the TUI panel track stands, the state of the one in-progress branch, and what must be true before the alternate screen becomes the default |
REA-OPENCODE-TUI.md |
Reverse-engineered opencode TUI study: OpenTUI engine evidence, settings/command registries, sidebar slots, tabs status, and fit for MSO |
TUI-PANEL-CONTRACT.md |
The invariants a TUI change must not break, each with the failure that proved it: byte-identical defaults, no setting without a consumer, and why updatedAt is not an activity signal |
AGENT-ACTIVITY-SIGNAL.md |
Why a live session activity signal cannot be derived today, the evidence per code path, and the smallest server change that would enable it |
ORGANIZATION.md |
Owner-private organization units, seats, reporting lines, live agent bindings, and workflow routing references |
MCP-HOW-TO.md |
Bahasa Indonesia onboarding: client connection, prompts, workflow, project MCP and troubleshooting |
MCP.md |
MCP/OAuth internals and full external-tool security model |
A2A.md |
A2A v1 peer discovery, delegation/task lifecycle, CLI/MCP surface and trust boundary |
LOCAL-AGENTS.md |
Native same-host session presence, identity, mailbox delivery, TUI/tools/API and isolation |
SUBAGENTS.md |
Same-session foreground isolated worker lifecycle, context, tool authority and limits |
AGENT-PROFILES.md |
Saved Owner profiles, exact model binding, revision history and bounded read tasks |
MCP-FEATURE-IMPLEMENTATION.md |
Stepwise reverse-engineering workflow for MCP tools, trusted skills and project capabilities |
CHATGPT-PLUGIN.md |
ChatGPT custom MCP app setup, diagrams, scopes and file bridge |
PLUGIN-REFERENCE-MATRIX.md |
Living cross-plugin architecture matrix, audit decisions and candidate generic primitives |
CONNECTED-APPS.md |
Server-persisted external app connections, dynamic shell windows and lifecycle boundaries |
MANAGED-APPS.md |
Hermes/OpenClaw/9Router lifecycle, jobs, backup/restore and dashboard access boundary |
MANAGED-DOMAINS.md |
Automatic app DNS/tunnel routes, named Integrations, HTTPS verification and durable setup recovery |
HERMES-INTEGRATION.md |
Hermes-specific integration behaviour |
OPENCLAW-INTEGRATION.md |
OpenClaw-specific integration behaviour |
9ROUTER-INTEGRATION.md |
9Router immutable Docker ownership, loopback default and explicit dashboard exposure |
ORCA-INTEGRATION.md |
Orca web bridge readiness/identity contract, private state root and shared-runtime operator limits |
MODELS-INTEGRATION.md |
Alfa BYOK/custom/Codex provider model |
SLICE-CATALOG.md |
Current slice/AppShell feature inventory |
FRONTEND-CONTRACT.md |
Required UI/UX implementation rules for every feature, state lifetime, responsive layout, actions, accessibility and evidence |
UI-UX-FOUNDATION.md |
Dynamic shell, feature and UI-kit spacing, workspace geometry and per-window feature state |
TROUBLESHOOTING.md |
Symptom → cause → supported recovery |
FAQ.md |
Product/security/operator boundaries |
../SECURITY.md |
Security posture and vulnerability reporting |
../CONTRIBUTING.md |
Contributor setup and gates |
../CLAUDE.md |
Repository/operator implementation rules |
../skills/README.md |
Skill catalog, trust and discovery semantics |
Design contracts awaiting implementation
| Document | Status |
|---|---|
EXTERNAL-FEATURE-CONTRACT.md |
v1 target contract for modular shell/external features; implementation status is explicit |
TENANT-PREVIEW.md |
Disabled identity and scoped-memory runtime; live activation and full production isolation remain unimplemented |
TENANT-PERSISTENCE.md |
Source-only durable registry, scoped memory and audit; transactions, recovery limits and operational gates |
TENANT-ROLLOUT.md |
Finite engineering, operational approval and acceptance gates for shared-gateway multiuser deployment |
NATIVE-HOST-RELAY.md |
Requested native macOS/Windows hosts and shared no-domain MCP relay; implementation and verification gates remain explicit |
Generated/current records
| Document | Source |
|---|---|
CLI.md |
node scripts/gen-cli-docs.mjs / bin/mso --help |
CHANGELOG.md |
Git subjects via scripts/gen-changelog.mjs; never hand-edit |
COMPARISON.md |
docs/comparison-data.json via scripts/gen-comparison.mjs; never hand-edit |
PROGRESS.md |
Hand-written shipping rationale; newest first; canonical historical WHY |
TESTING-HANDOFF.md |
Factual manual/automated test handoff; observed issues and resolved/remaining verification, never architecture authority |
Historical point-in-time documents
These are retained because they contain design reasoning or finding numbers referenced by source comments. They are not current implementation contracts.
| Document | Status |
|---|---|
EXTERNAL-FEATURE-AUDIT-2026-09-27.md |
Registry/manifest/embed source audit at 330d0cb4; migration phases remain pending |
9ROUTER-MSO-APPHOST-RFC.md |
Historical app-host proposal; current setup/authentication contract is in MANAGED-DOMAINS and MANAGED-APPS |
AUDIT-2026-08-24.md |
Fable/Ultracode + dynamic security audit at its named commit; includes reproduced findings and release remediation |
AUDIT-2026-06-11.md |
Archived audit at its named date; later fixes changed many findings |
R10-GLASS-AUDIT.md |
Point-in-time antislop R-10 glass dose audit; current rule compliance lives in the code |
SHELL-FIDELITY-PLAN.md |
Historical design backlog/baseline; current shell design lives in code + architecture |
DRAWER-MENU-BYOK-PLAN.md |
Historical implementation plan; phases A–C and OpenAI Codex D1 shipped; remaining ideas are not promises |
OS-SETTINGS-ICON-UPDATE-DEVICE-PLAN.md |
Open plan, not implemented: auto-update opt-in, stable channel, Devices/Server merge, account icon picker, upload→WebP/ICO pipeline |
HANDOVER.md |
Start here for a new session: current repo state, what shipped, what was cancelled and why, the numbers that decide things, and the traps that cost time |
MSO-MIGRATION.md |
Phased Next→Svelte/Bun/Rust plan with its phase table: 0/1/2 shipped, 3/4 cancelled by measurement, 6 no-go, 5 not started. Authoritative over BENCH-AFTER.md |
BENCH-BEFORE.md |
Measured baseline, 2026-10-29: bundle, LCP/TBT, hydration chain, per-app open cost and code surface, captured on an idle host as the "before" side of the Phase 4/5 comparison |
BENCH-AFTER.md |
Measured comparison, 2026-10-30: the same eight measures two days later, side by side, including the deltas that are zero and the ones inside the noise band |
MIGRATION-GATE-PLAN.md |
Executed migration-gate work: which §3 items landed with measurements, which were blocked and why, and the lib/ framework-import debt |
Marketing/demo collateral
These files are copy/script drafts, not technical authority:
When a current-reference doc and one of these drafts disagree, update the draft or follow the current-reference doc.
Reset and removal
Maintenance, factory reset and clean uninstall · Extended workspace reference
- Session screenshots and temporary artifacts — private per-session paths, MCP/CLI reads, quotas and safe retention.
Independent CI pilot
SELF-HOSTED-CI.md is the current manual-runner reference and Actions migration gap matrix. It does not claim completed security parity or deployment automation.