Obsidian iOS Real-Device Smoke

Core Rules

• Do not claim iOS real-device validation for a plugin-asset change unless the current build was written to the iCloud Obsidian test vault, all copied assets matched dist, the loaded plugin identity matched that artifact, and the affected behavior was observed on the connected iPhone.
• If only a vault-resident smoke runner changed, sync and hash-check that runner, then re-evaluate it in the existing page. Do not redeploy or reload the plugin when dist and plugin runtime assets did not change.
• Treat Safari Web Inspector as DOM/CSS/console/network evidence, not touch automation. Use iPhone Mirroring or the user for real touch interaction.
• Discover and use the available native UI tools for Safari and iPhone Mirroring; do not require or install a legacy skill by name. If no tool can operate the real target, report the blocked evidence instead of substituting emulation. For Safari Console JavaScript, use a real paste/input event and verify the command entered Console history or produced its sentinel before counting execution. Accessibility set_value may change the displayed editor text without creating executable Web Inspector input, so it is never submission evidence by itself.
• Treat an explicit user request to run iOS real-device smoke as authorization for plugin-asset deployment through make deploy-icloud (or its current-build reuse target), or the existing scoped preparation/sync command for a runner-only change, targeting only the iCloud test vault. Do not ask again. If the user requested only planning, review, or inspection without a real-device run, do not deploy or sync.
• Choose the smallest lane that covers the latest runtime delta and stop once that lane has sufficient current-build evidence.
• Judge a targeted canary against its current Decision/Spec/SDD contract, not a stricter topology borrowed from a deterministic Desktop fixture. In particular, a live Recovery canary must accept an intended reranker fail-open followed by recovery_relaxed / skipped / not_eligible when the standard attempt already returned usable evidence; require an actual relaxed retry only when the canary explicitly owns a deterministic valid-none/partial fixture. Likewise, lexical unavailable / feature_disabled / generation absent to ready / generation present with an advanced maintenance epoch is a valid first-generation transition. Preserve a raw runner false-negative, classify its product observations independently, and do not spend another device attempt solely to satisfy the invalid assertion.
• Before spending a submitted Console execution on a cancellation-only runner, require its post-load Memory plan to be ready / none. A fresh plugin reload with persisted rollout flags off may expose the existing lexical generation as feature_disabled; temporarily turning flags on then correctly requires preparation and is not a runnable cancellation precondition. Do not convert that setup-only BLOCKED into a product failure or repeat the command.
• Give every submitted attempt a unique result filename and copy/hash its raw receipt before the next attempt. Never let a retry overwrite the only device evidence. If historical tooling already overwrote a receipt, disclose that limitation and distinguish any retained projection from the unavailable raw file; do not describe the projection as immutable full evidence.
• Apply slice-scoped invalidation after a diagnostic-only repair. When an immutable real-device receipt already proves cancel request/observation, late discard, zero accepted-after-cancel, stable index identity and queue release, a successor that changes only terminal classification may close through exact new-artifact load identity plus a regression using the real wrapped AbortError shape. Do not rebuild Memory or spend another Provider/device attempt solely to replay the unchanged Worker/queue behavior.

Change Lanes

Lane	Use when	Required iPhone evidence	Skip
geometry-only	CSS/layout/position/size changed; gesture and action routing did not	Asset match, one reload, one real gesture that exposes the UI, one Mirroring screenshot, one Inspector geometry/hit-test probe, fresh errors	Hold threshold, movement cancel, exactly-once action checks, provider flows, unrelated orientations
interaction	Touch, pointer, keyboard, focus, dismissal, or action routing changed	Geometry basics plus the affected real gestures, cancel path, and exactly-once behavior	Unrelated Pagelet/provider/cross-surface flows
broad/release	Shared mobile runtime, multiple surfaces, packaging, or release gate	The declared cross-surface and orientation matrix	Anything explicitly outside the release scope

Do not promote a geometry-only delta into an interaction or release run merely because more evidence is available. Record unrequested orientations or devices as NOT TESTED residuals.

Device Preflight And Retry Budget

Before deployment or GUI control, confirm:

• the user requested real-device smoke;
• the local prerequisite covers the same runtime source state;
• iPhone Mirroring is connected and the device/Mac are unlocked;
• Obsidian is on the iCloud test vault and intended note;
• Safari Develop exposes the current Obsidian target.

Allow one unlock/reconnect/reattach sequence, or about three minutes, when a device or Inspector prerequisite is missing. Then report BLOCKED; do not keep hunting through menus or coordinates. After a reload, allow at most one Inspector reconnect. If body.in-progress persists, preserve the manual path and stop.

Use one submitted Console probe that returns all required assertions. Allow at most two execution attempts before reporting partial evidence or BLOCKED; submitted-but-ambiguous commands count as attempts. Do not spend additional attempts recovering focus, autocomplete, or stale element indices. Re-read app state after state-changing UI actions or window/DOM changes, and never reuse stale element indices; batch stable prompt entry, execution, and result capture where possible.

Workflow

Prerequisite: complete obsidian-test-vault-smoke at least at app-runtime tier.

1. Bound the changed surface:

git status --short --branch
git diff --stat
git diff --name-only

2. Reuse automated local checks with verified inputs/results under AGENTS.md Validation Planning And Reuse. Complete missing or invalidated Local Validation Gate checks, counting checks covered by the selected enclosing command and supplementing the runtime <style> / innerHTML / outerHTML source scan. Treat rg exit code 1 with no output as PASS and inspect every match. Reuse observed local app smoke only within the same turn while its target/runtime state is unchanged; otherwise renew the affected local app proof. For broad UI/runtime changes, retain make deploy (or its eligible current-build target) and local app smoke before device validation.

3. Sync only the changed artifact after the user requested real-device smoke:

• Runner-only change: use that runner's preparation/sync workflow, compare its destination hash with the source, and keep the current plugin instance and Inspector session. Skip plugin deployment and reload.
• Plugin runtime or CSS change: deploy the current plugin assets:

make deploy-icloud

If the required checks already passed for the current changes (for example, during desktop smoke), use make deploy-icloud-current to verify and copy the same current production build. It checks build identity, not test results; follow AGENTS.md Local Deployment reuse conditions. If both destinations are authorized together, make deploy deploy-icloud shares full validation once.

This writes only plugin assets under:

~/Library/Mobile Documents/iCloud~md~obsidian/Documents/test/.obsidian/plugins/personal-assistant

Request escalated permissions if the sandbox blocks this expected iCloud write.

4. For a plugin-asset change, verify all copied assets deterministically:

ICLOUD_PLUGIN_DIR="$HOME/Library/Mobile Documents/iCloud~md~obsidian/Documents/test/.obsidian/plugins/personal-assistant"
for file in main.js manifest.json manifest-beta.json styles.css; do
  if ! cmp -s "dist/$file" "$ICLOUD_PLUGIN_DIR/$file"; then
    printf 'MISMATCH %s\n' "$file"
    exit 1
  fi
  printf 'MATCH %s\n' "$file"
done

Stop with FAIL if any source or destination file is missing or differs. For a runner-only change, verify only the runner source/destination identity and confirm the already-loaded plugin identity is still current; do not manufacture a plugin lifecycle transition as extra evidence.

5. Load the changed artifact on iPhone using the cheapest sufficient path:

• Runner-only: re-evaluate the synced runner in the existing page. Do not reload the plugin, vault, or App.
• Plugin assets: after byte comparison passes, reload only personal-assistant first, using Obsidian's plugin reload lifecycle. Capture the previous plugin object before reload.
• Before product assertions, require the new plugin object to exist and differ from the previous object, then call getLoadedPluginBuildIdentity() and require no blocker plus loadedPluginArtifactSha256 equal to the matched iCloud main.js SHA-256. Also preserve its load timestamp/version in the evidence.
• If plugin reload is unavailable or identity proves it did not load the matched artifact, allow exactly one vault reload or normal App reopen as a fallback, not both. Recheck loaded identity before continuing.
• Force-quit Obsidian from the iPhone background only when the declared evidence target is process startup, OPFS persistence, or complete unload, or when the reload path above has been proven ineffective by identity evidence. Record that reason; stale-state suspicion alone is insufficient.
• Stop repeated attempts if Obsidian stalls at body.in-progress.
• Do not classify an unresponsive tap as a product bug until asset comparison passed and the iPhone loaded the new build.

6. Use the available native UI tools to inspect Safari Web Inspector attached to the iPhone target named like -- Obsidian -- localhost and to operate iPhone Mirroring when available.

• If the target, device trust, Develop menu, or Web Inspector connection is unavailable, report BLOCKED; do not guess from desktop/mobile emulation.
• Use paste for long Console probes and verify history/output; do not submit a value populated only through Accessibility set_value.

For a Pagelet Action Ring geometry-only change, expose the Ring with one real long press and capture the Mirroring screenshot. If it auto-closes before Inspector measurement, let this single probe reopen it through the production keyboard/context-menu path solely for DOM geometry. Do not count that synthetic reopen as touch evidence.

(()=>{
  const trigger=document.querySelector(".pa-pagelet-pet-trigger");
  if(!trigger){
    console.log('PA_ACTION_RING_LAYOUT {"error":"pet-missing"}');
    return;
  }
  if(!document.querySelector(".pa-pagelet-action-ring")){
    trigger.dispatchEvent(new KeyboardEvent("keydown",{key:"ContextMenu",bubbles:true,cancelable:true}));
  }
  requestAnimationFrame(()=>requestAnimationFrame(()=>{
    const items=[...document.querySelectorAll(".pa-pagelet-action-ring-item")];
    const viewport=window.visualViewport;
    const box=viewport
      ? {left:viewport.offsetLeft,top:viewport.offsetTop,right:viewport.offsetLeft+viewport.width,bottom:viewport.offsetTop+viewport.height}
      : {left:0,top:0,right:innerWidth,bottom:innerHeight};
    const read=(el)=>{
      const r=el.getBoundingClientRect();
      const hit=document.elementFromPoint(r.left+r.width/2,r.top+r.height/2);
      return {
        text:el.textContent?.trim(),
        rect:[r.left,r.top,r.width,r.height].map((n)=>+n.toFixed(1)),
        hitSelf:hit===el||el.contains(hit),
      };
    };
    const rows=items.map(read);
    console.log("PA_ACTION_RING_LAYOUT "+JSON.stringify({
      trigger:read(trigger),
      actions:rows,
      count:rows.length,
      sameTop:rows.length===3&&rows.every((x)=>Math.abs(x.rect[1]-rows[0].rect[1])<=1),
      increasingLeft:rows.every((x,i)=>i===0||x.rect[0]>rows[i-1].rect[0]),
      nonOverlapping:rows.every((x,i)=>i===0||x.rect[0]>=rows[i-1].rect[0]+rows[i-1].rect[2]),
      min44:rows.every((x)=>x.rect[2]>=44&&x.rect[3]>=44),
      insideViewport:rows.every((x)=>x.rect[0]>=box.left&&x.rect[1]>=box.top&&x.rect[0]+x.rect[2]<=box.right&&x.rect[1]+x.rect[3]<=box.bottom),
      allHitSelf:rows.every((x)=>x.hitSelf),
    }));
  }));
})()

Require count=3, all boolean assertions true, the intended label order, and a matching visible screenshot. If gesture code did not change, stop there after fresh error capture; do not repeat movement-cancel or exactly-once checks.

Probe real Obsidian chrome before tuning plugin UI. Adjust thresholds for device and orientation:

(()=>{const q=[...document.querySelectorAll('button,.clickable-icon,.pa-pagelet-pet')].filter(e=>{const r=e.getBoundingClientRect();return r.top<130&&r.left<430&&r.width>10&&r.height>10}).map(e=>{const r=e.getBoundingClientRect(),s=getComputedStyle(e),svg=e.querySelector('svg'),p=svg&&svg.querySelector('path,line,polyline,rect,circle');return{cls:String(e.className),title:e.title,aria:e.getAttribute('aria-label'),rect:[r.x,r.y,r.width,r.height].map(n=>+n.toFixed(1)),color:s.color,svg:svg&&{rect:[svg.getBoundingClientRect().x,svg.getBoundingClientRect().y,svg.getBoundingClientRect().width,svg.getBoundingClientRect().height].map(n=>+n.toFixed(1)),viewBox:svg.getAttribute('viewBox'),w:svg.getAttribute('width'),h:svg.getAttribute('height')},path:p&&{tag:p.tagName,stroke:p.getAttribute('stroke'),sw:p.getAttribute('stroke-width'),cssStroke:getComputedStyle(p).stroke,cssSw:getComputedStyle(p).strokeWidth}}});console.log('PA_ICON_PROBE '+JSON.stringify(q))})()

Use a null-safe focused pet/system-button comparison. .sidebar-toggle-button.mod-left is an Obsidian internal class and may be absent after an app update:

(()=>{let l=document.querySelector('.sidebar-toggle-button.mod-left'),p=document.querySelector('.pa-pagelet-pet');let f=e=>{if(!e)return null;let r=e.getBoundingClientRect(),s=e.querySelector('svg'),a=s?[...s.querySelectorAll('path,rect,line,polyline,circle')]:[];return{r:[r.x,r.y,r.width,r.height].map(n=>+n.toFixed(1)),svg:s&&[s.getAttribute('viewBox'),+s.getBoundingClientRect().width.toFixed(1),+s.getBoundingClientRect().height.toFixed(1)],sw:[...new Set(a.map(x=>(x.getAttribute('stroke-width')||getComputedStyle(x).strokeWidth)))],csw:[...new Set(a.map(x=>getComputedStyle(x).strokeWidth))].slice(0,6)}};console.log('PA_MIN '+JSON.stringify({left:f(l),pet:f(p)}))})()

Pitfalls and Stop Conditions

• Measure SVG geometry and stroke width; do not infer apparent weight from screenshots alone.
• Verify asset matches and reload before treating stale iCloud behavior as a regression.
• Treat Reduce Motion as a hypothesis until verified.
• Do not generalize phone placement to iPad without separate evidence.
• Stop automated reloads if body.in-progress persists and preserve the user's manual testing path.
• Do not escalate runner refresh into plugin reload, or plugin reload into vault/App restart, without the identity evidence required above.
• Stop after the selected lane's evidence is complete; do not expand scope to improve an already sufficient proof.
• Stop before writing outside the iCloud test plugin directory or the declared runner/fixture paths, and never delete vault data.

Output

iOS real-device smoke:
- Lane: `<geometry-only/interaction/broad-release>`
- Local prerequisite: PASS/FAIL
- Reused evidence: `<same-state evidence or none>`
- Device preflight: PASS/BLOCKED
- Artifact sync/deploy: PASS/FAIL/NOT_RUN
- Artifact evidence: `<runner-only: source/destination runner hash + unchanged current plugin identity | plugin-assets: four MATCH lines>`
- Load path: `<runner re-evaluation / plugin reload / single fallback / justified process restart>`
- Loaded identity: `<runner-only: unchanged loaded SHA/version | plugin-assets: instance transition + loaded SHA/version/load timestamp>`
- iPhone observation: PASS/FAIL/BLOCKED - `<Mirroring / Inspector / user>`
- Findings:
  - PASS: `<path>` - `<observed behavior>`
  - FAIL: `<path>` - `<issue and user impact>`
  - BLOCKED: `<path>` - `<external blocker>`
- Inspector evidence: `<rects, CSS, console, DOM state>`
- Explicit skips: `<unaffected gestures, surfaces, orientations, or devices>`
- Stop point: `<sufficient evidence or retry limit reached>`
- Residual risk: `<untested paths or stale-state concern>`

Separate “deployed and asset-matched” from “observed on iPhone.” Never promote the former into a real-device PASS.

Related Skills

• Use obsidian-test-vault-smoke first at app-runtime tier or higher.
• Use available native UI tools for Safari and iPhone Mirroring UI control.
• Use obsidian-community-check only for an authorized hosted community scan.
• Use personal-assistant-review for code-level review gates.