docs: improve skill mental model wording

Author: thymikee

skills/agent-device/SKILL.md

@@ -12,17 +12,23 @@ Use this skill as a router.
 - Open-ended bug hunt with reporting: use [../dogfood/SKILL.md](../dogfood/SKILL.md).
 - Pass/fail QA from acceptance criteria: stay in this skill, start with [references/bootstrap-install.md](references/bootstrap-install.md), then use the QA loop in [references/exploration.md](references/exploration.md).
 
-Default route inside this skill: bootstrap -> exploration -> optional debugging or verification. Open the macOS and remote-tenancy references only for those exceptions.
+Default route inside this skill: bootstrap -> exploration -> optional debugging or verification. Open the macOS reference only for host Mac desktop work. Open the remote-tenancy reference only for remote daemon HTTP or lease flows.
+
+## Mental model
+
+- First choose the correct target and open the app or session you want to work on.
+- Then inspect the current UI with `snapshot -i` and pick targets from the actual UI state.
+- Act with `press`, `fill`, `get`, `is`, `wait`, or `find`.
+- Re-snapshot after meaningful UI changes instead of reusing stale refs.
+- End by capturing proof if needed, then `close`.
 
 ## Decision rules
 
-- Use `app` sessions to act.
-- Use `frontmost-app`, `desktop`, and `menubar` mainly to inspect until helper interaction parity exists.
 - Use plain `snapshot` when you need to verify whether text is visible.
 - Use `snapshot -i` mainly for interactive exploration and choosing refs.
 - Use `fill` to replace text.
 - Use `type` to append text.
-- Prefer selector or `@ref` targeting over raw coordinates.
+- Prefer `@ref` or selector targeting over raw coordinates.
 - Keep the default loop short: `open` -> explore/act -> optional debug or verify -> `close`.
 
 ## Choose a reference
@@ -32,4 +38,4 @@ Default route inside this skill: bootstrap -> exploration -> optional debugging
 - Need logs, network, alerts, permissions, or failure triage: [references/debugging.md](references/debugging.md)
 - Need screenshots, diff, recording, replay maintenance, or perf data: [references/verification.md](references/verification.md)
 - Need host macOS desktop behavior or surface differences: [references/macos-desktop.md](references/macos-desktop.md)
-- Need remote daemon transport or lease admission: [references/remote-tenancy.md](references/remote-tenancy.md)
+- Need to connect to a remote `agent-device` daemon over HTTP or use tenant leases: [references/remote-tenancy.md](references/remote-tenancy.md)

skills/agent-device/references/macos-desktop.md

@@ -15,7 +15,7 @@ Open this file only when `--platform macos` is involved or the task needs `front
 
 ## Most common mistake to avoid
 
-Do not treat desktop surfaces like normal action surfaces. Use `app` sessions to act, and use `frontmost-app`, `desktop`, or `menubar` mainly to inspect until helper interaction parity exists.
+Do not treat every macOS surface the same. Use the normal `app` surface when you want to act inside one app. Use `frontmost-app`, `desktop`, or `menubar` mainly to inspect what is visible before switching back to `app` for most interactions.
 
 ## Canonical loop