First Mission (guided onboarding experience)
The short coached experience that runs right after a new user finishes setup: a read-only health check across performance, security, privacy, drive health and hardware, then the worthwhile fixes one at a time, each explained and approved by you.
On this page
What it is
The First Mission is a short, coached experience that runs automatically right after a new user finishes onboarding setup on Windows, macOS or Linux. It gives the user's computer a safe, guided tune-up — a read-only health check across performance, security, privacy, drive health, and hardware — then walks them through the worthwhile fixes one at a time, each explained in plain language and approved (or skipped) individually. Personal files are never touched without an explicit, per-item OK, and it finishes with a plain-English before/after report.
It fires at most once per install (when neither firstMissionCompleted nor firstMissionDismissed is set in the user's Omniscio settings), and it does not run again once the user has either completed or dismissed it. It is offered on Windows, macOS and Linux — MISSION_REGISTRY lists all three — though the scripted scan it runs is written for Windows and macOS today (see Platform support).
Where to find it
When it appears
The First Mission launches immediately after the user clicks Create Project in the final step of the onboarding wizard (the Project step, where they pick a folder). Instead of opening a blank session, Omniscio opens a real working session: the agent introduces itself and asks one quick yes/no — whether it can run a read-only check of the system — then runs the scan once the user agrees (each system command still goes through the app's normal command-approval flow). It is also the "Tidy my computer" agent the onboarding tour spawns as the user's first real session. The session is otherwise indistinguishable from a regular session — it lives in the sessions sidebar, uses the active Claude account, and can be archived, paused, or otherwise managed like any other session.
If the host platform has no missions in MISSION_REGISTRY, or if the mission has already been completed or dismissed, the Project step falls back to a plain blank session exactly as it did before the First Mission feature existed.
There are two ways in: the automatic launch above, and the Agent Missions hub (see How to replay it), which starts the same mission on demand and ignores whether it was completed or dismissed.
Two further entry points were built and are currently switched off — a blank-session nudge ("New here? Let me show you something useful" with a Run a guided mission button) and a "Ready to go" replay button in the main-panel empty state — together with the mission picker, all behind MISSION_UPSELL_SURFACES_ENABLED in src/shared/missions/registry.ts (retired 2026-08-26 pending the video-led onboarding cadence). On a normal build those buttons do not render. Because the AUTOMATIC launch fires only once, an install that has already completed or dismissed the mission opens a plain blank session on every later onboarding run — the hub is how you get it back.
How it behaves
The flow
The mission is a normal conversation with a real working agent (the scripted card beats were retired 2026-08-11 — the mission is a real working agent that responds, not a canned card script):
Ask permission first — the agent introduces itself in a sentence or two and asks one yes/no: may it run a read-only check of the computer? Nothing happens until the user agrees; a "no" gets a polite sign-off and a full stop.
Scan quietly — on a yes, it sets up a safety net first (a Windows System Restore point, or verifies Time Machine on Mac), then runs a read-only scan across about a dozen areas — disk junk, startup programs, bloatware, updates, power, privacy, security, drive health, backups, and account security. The user sees nothing until the findings are ready.
Findings, then one fix at a time — the agent shows a short, plain-language overview (one line per area with the key number), then walks through the worthwhile fixes individually. Each is explained — what it is, why it matters, what changing it does — and the user approves or skips it item by item, never a blanket "do everything."
Safe by default, with undo — personal files stay off-limits unless approved item by item, every change is reversible and recorded, and a decline at any point gets one warm sentence and a full stop — never persuasion. It closes with a plain-English before/after report.
Trust model
The mission is designed around explicit consent at every step:
- The scan is strictly read-only. No files are touched, and the user agrees to it up front.
- A safety net comes first. Before any change, the agent creates a System Restore point on Windows (or verifies Time Machine on Mac), so anything can be rolled back.
- Every change is approved item by item. Nothing is executed on a blanket "yes," personal folders (Documents, Desktop, Pictures, work folders, etc.) stay off-limits without explicit per-item consent, and each applied fix is reversible and recorded for the final report.
Because the agent runs under the app's normal command approval, every system command is individually visible and approvable — there is no hidden execution path.
How to replay it
If a user dismissed the mission with "Maybe later", or completed it and wants a fresh tune-up, replay it from the Agent Missions hub — the Missions row in the sidebar (missionsSidebarEnabled, on by default):
- Open Missions and find the Tidy my computer card (that is the First Mission). Press Start.
- Omniscio opens a new session in the Missions project with the mission's prompt pre-loaded, and the run is identical to the onboarding one. This path does not read
firstMissionCompleted/firstMissionDismissed, so it works however the first run ended. - The "Ready to go" button and the MissionPickerModal from earlier builds are still hidden behind
MISSION_UPSELL_SURFACES_ENABLED(off) — the hub card is the live manual entry.
Platform support
The mission is offered on Windows, macOS and Linux, because the registry is the gate: MISSION_REGISTRY in src/shared/missions/registry.ts lists platforms: ['win32', 'darwin', 'linux'] for every mission, and UI surfaces gate on hasMissionsForPlatform(getHostPlatform()), which is therefore true on all three. A host the registry does not list skips the mission and opens a plain blank session exactly as before.
The scripted tune-up is Windows and macOS only today: resources/first-mission/mission-prompt.md carries Windows detection (Get-CimInstance, Win32_*), Mac detection, and a System Restore / Time Machine safety net — it has no Linux branch, no uname, systemctl or distro handling. On Linux the mission still launches, but its health check is whatever the agent manages to improvise from the shared instructions rather than a Linux-specific scan. Treat a Linux run as best-effort, and expect that gap to close in the prompt rather than in the registry.
Settings flags
Omniscio tracks the mission state in two boolean settings:
| Setting | What it means |
|---|---|
firstMissionCompleted |
The mission ran to completion (the tune-up finished, or the user stopped after the findings) |
firstMissionDismissed |
User opted out (declined the mission, or the blank-session nudge's "don't show again" link) |
Both default to false. They gate the automatic mission launch from ProjectStep: when either is true, ProjectStep opens a normal blank session instead of the mission. They do not gate the Agent Missions hub's Tidy my computer card, which starts the mission whenever the user asks — that is the replay path, not a flag.
Related
- first-time-setup.md — the onboarding wizard that precedes the First Mission
- question-widget.md — the marker-driven card pattern the mission mirrors
- start-a-new-session.md — the normal session launch flow the First Mission extends
- default-claude-project.md — the auto-created
~/Claudeproject where replay missions often run - ai-workflow-integration-planner-mission.md — the other guided mission the Agent Missions hub can start: an interview about your work that returns concrete AI-integration ideas.
Last verified 2026-09-23