create-mobile-app

📋 Shared instructions: shared-instructions.md — read first. Covers safety guardrails, memory bank usage, preferred-environment policy, connector-first rule, Windows CLI compat, command-failure handling.

Create Power Apps Code App (Native)

Top-level orchestrator. Owns the user-visible flow; delegates planning to the native-app-planner agent and per-domain mutation to dedicated /add-* skills.

Workflow

0. Resume check + fresh-template gate → 1. Prerequisites → 2. Gather requirements → 2b. Requirements discovery → 2c. Plan preview (rough cost + abort gate) → 3. Plan (planner agent + 4 gates) → 4. Auth & environment → 5. Prepare existing template → 6. npx power-apps init → 6.5 verify npm install → 6.5b SafeAreaProvider gate (always runs, idempotent) → 6.6 scaffold tsc smoke check → 6.7 seed memory bank → 6.85 Offline profile (always asked) → 7. Auth config → 8. Apply data model → 9. Apply native capabilities → 9a. Install planned JavaScript dependencies → 9b. Design system → 10. Add connectors → 10b. Wire navigation layout → 11. Build screens (parallel) → 11.4 Stylistic fix sweep → 12. Start Metro (npx expo start) → 12.5 Optional debug handoff → 13. Summary

---

Fresh-template working-directory mode

This skill assumes the user already has a fresh microsoft/power-platform-skills/plugins/mobile-apps/template#main template materialized with degit in the target working directory and has already run npm install there. The skill turns that fresh template into an app; it does not clone, degit, or copy a template itself.

Fresh template required. If the working directory is not a template, or if it already looks like an app created by this skill, STOP and tell the user to materialize a fresh microsoft/power-platform-skills/plugins/mobile-apps/template#main template with degit into a new folder, run npm install, then rerun /create-mobile-app --working-dir <fresh-template-dir>.

Use these markers:

State	Detection	Action
Fresh template	package.json, app.config.js, auth.config.json, tamagui.config.ts exist; node_modules/expo exists; memory-bank.md, native-app-plan.md, .datamodel-manifest.json, and generated Dataverse services are absent	Proceed.
Template not installed	Fresh-template files exist but node_modules/expo is absent	STOP: ask user to run npm install in the template folder, then rerun. Do not provision ADO npm tokens here.
Already-created app	memory-bank.md, native-app-plan.md, .datamodel-manifest.json, or src/generated/services/*.ts exists	STOP: this is not a fresh create target. Ask user to materialize a fresh template folder with degit.
Not template	Required template files are missing	STOP: ask user to materialize microsoft/power-platform-skills/plugins/mobile-apps/template#main into the working directory with degit and run npm install.

This gate is intentionally simple: /create-mobile-app creates a new app from a fresh template. It does not adopt, repair, resume, or overwrite an already-created app.

---

TypeScript Gate Policy — no quality compromise

tsc is a phase gate, not a reflex after every tiny edit. The app may not advance past a gate until TypeScript is clean.

Required gates:

• Scaffold gate: Step 6.6 after existing-template preparation, npx power-apps init, and dependency verification.
• Dataverse/generated-services gate: immediately after Step 8 returns and generated services/models are refreshed.
• Navigation/skeleton gate: after Step 10b layouts and Step 10.8 shared code/skeletons are written, before Step 11 builders launch.
• Screen-wave gate: after each Step 11 screen-builder wave returns, before launching the next wave.
• Final gate: before Step 12 starts the dev server.

When a gate fails:

1. Capture the full tsc --noEmit output once.
2. Classify errors by root cause (for example: generated model names, service option shapes, invalid UI props, typed percentage values, create/update payload typing, missing imports).
3. Repair in a batch.
4. Re-run the same gate once after the batch.
5. Continue only when the gate is clean, or stop/block according to the retry policy.

Do not run full-app tsc after every microscopic local edit inside the same repair pass. That is slower and encourages line-by-line patching. Batch root-cause fixes, then re-run the gate. This is a speed improvement only; it does not lower the quality bar.

Hard stops:

• Do not launch data-source work from a broken scaffold gate.
• Do not launch screen-builders from broken generated services, layouts, shared code, or skeletons.
• Do not launch wave N+1 until wave N passes its tsc gate.
• Do not start the dev server until the final gate is clean.
• Do not hide approved capability failures behind mocks or TODOs just to satisfy tsc.

---

Step 0 — Resume check + fresh-template gate

Telemetry checkpoint: template_gate

If $ARGUMENTS includes a --working-dir (or the user names an existing directory), check whether <working_dir>/memory-bank.md exists.

• Bank present → read it. Identify the highest-numbered completed step. Inform the user:

  "Found existing project '' at <dir>. Steps 1– already completed (last update ). Resume from Step <N+1>?" Wait for confirmation. If the user says yes, jump to that step. Skip the wizard (Step 2) and re-use the values stored in the bank.

• Bank absent → fresh project. Continue to Step 1.
• Bank present but corrupted (missing required headings) → surface the parse error, ask the user whether to overwrite (lose history) or fix manually before proceeding.

The bank is the only resume mechanism. Do not infer resume state from package.json or node_modules/ — those can lie.

After the resume check, run the fresh-template gate from the section above. This is a create-only command:

• If memory-bank.md exists and the user confirms resume, resume as documented above.
• If any already-created-app marker exists and there is no approved resume path, STOP and tell the user to materialize a fresh template into a new folder with degit.
• If required template files are missing, STOP and tell the user to materialize microsoft/power-platform-skills/plugins/mobile-apps/template#main into the working directory with degit and run npm install.
• If node_modules/expo is missing, STOP and tell the user to run npm install in that template folder before rerunning this skill.

Do not silently copy a bundled template over the user's folder. A fresh plugins/mobile-apps/template template may contain placeholder power.config.json with an empty environmentId; Step 5 removes that placeholder immediately before Step 6 runs npx power-apps init.

After the fresh-template gate succeeds, or after the user confirms a resume, initialize the app identity before continuing:

node "${PLUGIN_ROOT}/scripts/lib/app-identity.js" "<working_dir>"

app-identity.js mints app.json expo.extra.telemetry.appInstanceId — a random per-project ID that lets usage telemetry tell this app apart from other apps built in the same session, and recognize it again in later sessions. It is idempotent, so a resume or re-run keeps the original ID. It contains no app name, path, or environment data. Commit it: it is the app's identity, not a per-machine cache.

Step 1 — Prerequisites

Telemetry checkpoint: prerequisites

Run all checks first — no point gathering requirements if the toolchain isn't ready.

Important: npm auth and Power Platform app auth are separate. The account used for npm install can be different from the account used by npx power-apps:

What	Uses	Typical account
npm install private feed access	npm/Azure Artifacts auth configured outside this skill	Account with feed Reader access
npx power-apps init, Dataverse, deploy	npx power-apps browser auth + az login --tenant <env-tenant> for Dataverse helper scripts	Power Platform environment account, often a test-tenant/admin account

Renewing npm feed auth does not sign the user into npx power-apps. If the Power Apps CLI prompts for browser auth later, that is expected and unrelated to the npm/ADO feed token.

Then run the checks:

node --version                                      # v22+
npm  --version                                      # v10+
az account show --query "user.name" -o tsv          # Azure CLI logged in (needed for Dataverse helper scripts)
git --version                                       # optional

Do NOT probe Xcode, Java, Android Studio, or CocoaPods here. This plugin's flow is plan → scaffold → code → local Expo dev server. Build + deploy (npm run build / npx power-apps push) is a separate user-driven step via the /deploy skill. Local native compile is the user's choice and lives outside this skill (run the platform-specific native command directly when needed). See shared/version-check.md — only the Always required tier matters here.

Missing	Action
Node < 22	STOP — instruct nvm install 22 && nvm use 22
az	STOP — instruct az login

Template-only rule: this skill no longer provisions npm feed tokens, PAT fallbacks, vendor fallbacks, or registry rewrites. The user must run npm install in the fresh template folder before invoking /create-mobile-app.

Capture target Power Platform environment for the remaining flow.

Source of truth for env selection: the generated power.config.json first, explicit environment ID second. In the normal template-folder flow, npx power-apps init runs first and writes the selected environment ID into power.config.json; read that ID and pass it to scripts/resolve-environment.js to resolve the Dataverse URL and tenant. If power.config.json is missing or has an empty placeholder environmentId, ask for an environment ID. A Dataverse URL is useful as a resolver fallback for existing apps, but it is not enough for npx power-apps init because init needs --environment-id.

Step	Source	When user is asked
0. power.config.json has environmentId	scripts/resolve-environment.js <environment-id>	Never — automatic after npx power-apps init
1. User supplies env ID	scripts/resolve-environment.js <environment-id>	Ask only if power.config.json is missing/empty or user wants a different env
2. User wants a different account	Follow shared-instructions standalone CLI auth handling	Only if resolution/token acquisition fails or user asks
3. User wants different env	Ask for another env ID and re-run resolver	Only if user selects "use a different environment" at Step 2
4. npx power-apps init -t MobileApp --display-name "$DISPLAY_NAME" --environment-id $ACTIVE_ENV_ID --non-interactive	Persists choice into power.config.json	Only when this skill owns the initial init path

TARGET_ENV="<environment-id-or-empty>"
if [ -z "$TARGET_ENV" ] && [ -f power.config.json ]; then
  TARGET_ENV=$(node -e "try { const id=require('./power.config.json').environmentId || ''; console.log(id); } catch { console.log(''); }")
fi
test -n "$TARGET_ENV" || { echo "✗ Environment missing. Provide an environment ID."; exit 2; }
ENV_JSON=$(node "${PLUGIN_ROOT}/scripts/resolve-environment.js" "$TARGET_ENV")
printf '%s\n' "$ENV_JSON" > .resolved-environment.json
ACTIVE_ENV_ID=$(node -e "const j=JSON.parse(process.argv[1]); console.log(j.environmentId || '')" "$ENV_JSON")
ACTIVE_ENV_NAME=$(node -e "const j=JSON.parse(process.argv[1]); console.log(j.displayName || j.environmentUrl || '')" "$ENV_JSON")
ACTIVE_ENV_URL=$(node -e "const j=JSON.parse(process.argv[1]); console.log(j.environmentUrl || '')" "$ENV_JSON")
ACTIVE_TENANT_ID=$(node -e "const j=JSON.parse(process.argv[1]); console.log(j.tenantId || '')" "$ENV_JSON")
test -n "$ACTIVE_ENV_ID" || { echo "✗ Environment ID missing. Provide the environment ID directly."; exit 2; }
echo "✓ Target env: $ACTIVE_ENV_NAME ($ACTIVE_ENV_ID)"
echo "✓ Target env URL: $ACTIVE_ENV_URL"
echo "✓ Target tenant: ${ACTIVE_TENANT_ID:-unknown}"

Orchestrator handling for exit 2: ask the user for their environment ID directly, then re-run the capture block above. Do not run npx power-apps init here; Step 6 owns initialization after the user confirms the target environment.

Stash $ACTIVE_ENV_ID, $ACTIVE_ENV_NAME, $ACTIVE_ENV_URL, and $ACTIVE_TENANT_ID for Step 2 (env confirmation), Step 6 (npx power-apps init), and Step 7 (auth.config.json tenant/environment cache). If parsing fails, ask for an environment ID again.

If resolve-environment.js cannot get tokens, run az login --tenant <env-tenant> in the foreground. If npx power-apps init later uses the wrong account, follow shared-instructions standalone CLI auth handling and retry once.

Step 1.7 — Detect publisher prefix

Detect the publisher prefix for the env's Default solution so the planner uses the correct prefix rather than assuming cr_.

Deferred execution: do not run the query at this point. Step 2b.4 first classifies the run as required or connector-only; only required runs execute the block below. Connector-only runs set $DETECTED_PUBLISHER_PREFIX = "" and make no Dataverse prefix query.

node "${PLUGIN_ROOT}/scripts/detect-publisher-prefix.js" "$ACTIVE_ENV_URL" --tenant-id "$ACTIVE_TENANT_ID"

Output is one line of JSON, e.g.:

{"prefix": "cr8142a", "source": "detected"}
{"prefix": null, "reason": "no token (run `az login --tenant <env-tenant>`)"}

The script queries the Default solution's publisher via: /api/data/v9.2/solutions?$select=uniquename&$expand=publisherid($select=customizationprefix)&$filter=uniquename eq 'Default'

A second solution name can be passed as a second argument if the env uses a different solution (defaults to 'Default').

Token tenant note: the script's getAuthToken discovers the env's tenant ID from the Dataverse HTTPS auth challenge and passes --tenant <env-tenant> to az, so detection works even when the active az identity is on a DIFFERENT tenant. If the user has not run az login --tenant <env-tenant> at any point, detection may return null.

Stash the result for Step 3 (planner spawn):

Output	Stash as	Behavior at Step 3
{"prefix": "cr8142a", ...}	$DETECTED_PUBLISHER_PREFIX = "cr8142a"	Pass to planner prompt as a fact: "Publisher prefix (detected from env): cr8142a_"
{"prefix": null, ...}	$DETECTED_PUBLISHER_PREFIX = "" (empty)	Pass to planner as: "Publisher prefix: NOT DETECTED — use placeholder cr_ and warn the user that Dataverse will normalize the actual prefix at create time."

Do NOT block on null detection — the user can still proceed; the Power Apps CLI normalizes prefixes when npx power-apps add-data-source runs. The detection step is purely to make the plan output accurate.

If the script exits non-zero (rare — should always exit 0 with prefix: null), treat it as the null case and continue.

Step 2 — Gather requirements

Skip questions the user already answered in $ARGUMENTS.

If the user gave no description, ask one open-ended question first:

"What would you like to build? Describe it in your own words — what it does, who uses it, and what problem it solves."

Then collect with AskUserQuestion (batch where possible):

Question	Default
App display name	derived from description
Target platforms	ios, android (multi-select, default both)
Aesthetic	minimal / playful / professional / matches existing brand
Target environment	Confirm <ACTIVE_ENV_URL> / <ACTIVE_ENV_ID> from Step 1.6, or choose "use a different environment" and provide another environment ID

App slug is auto-derived from the display name (slugify(displayName) — kebab-case, ASCII-only, strip non-alphanumerics). Do NOT ask the user; the derived slug is correct >95% of the time. Show the resolved slug as part of Step 2c's plan preview so the user can override via edit if needed.

Environment override branch: If the user picks "use a different environment", ask for the Power Platform environment ID via AskUserQuestion, then run scripts/resolve-environment.js again and refresh $ACTIVE_ENV_ID / $ACTIVE_ENV_URL / $ACTIVE_TENANT_ID.

App-name collision pre-flight. Once <displayName> is fixed, check the chosen env for a name collision:

npx power-apps list-codeapps --environment-id "$ACTIVE_ENV_ID" --json 2>/dev/null | grep -F "<displayName>" >/dev/null && \
  echo "COLLISION" || echo "OK"

If COLLISION, ask the user via AskUserQuestion:

"An app named <displayName> already exists in <ACTIVE_ENV_NAME>. Choose:

1. Pick a different name (recommended)
2. Delete the existing app in Maker portal — DESTRUCTIVE, asks confirmation outside this skill
3. Continue anyway (bg npx power-apps init will fail; you'll have to rename later — NOT recommended)"

Re-prompt for name if (1). If (2), send the user to Maker portal to delete the existing app, then re-run the collision check. Only proceed once collision is resolved.

If npx power-apps list-codeapps is unavailable in the installed CLI version, skip the pre-flight silently and continue.

Don't enter plan mode here — that's the planner agent's job in Step 3.

Step 2b — Requirements discovery

Goal: Turn the user's thin prompt into a confirmed feature brief before the planner runs. The planner agent receives this brief verbatim — richer input means better data model inference, accurate connector detection, and correct screen specs.

Step 2b.0 — Prompt richness scoring (decides which path to take)

Before asking anything, score the description on four signals. The score decides whether we ask a multi-select feature picker, a single confirmation, or skip the discovery question entirely.

Run this scorer mentally on <description> (the prompt the user gave with /create-mobile-app, plus any clarifying text from Step 2a). Count how many of the four trip:

Signal	Trips when
Word count	description has ≥ 60 words
Distinct nouns	description names ≥ 5 distinct domain nouns (people, things, documents, places — e.g. "inspector", "aircraft", "gate", "defect", "evidence")
Action verbs	description uses ≥ 3 workflow verbs from this set: log, track, submit, assign, notify, scan, upload, approve, verify, complete, capture, override, dispatch, review, sign
Domain phrase	description names a known industry domain — match against the industry table in shared/references/universal-patterns.md (airline, hospital, retail, manufacturing, field-service, finance, logistics, …) OR explicitly says "field operations" / "ground operations" / "site visit" / similar

Tier the result:

Score	Tier	What to do
4 / 4	auto-plan	Skip both questions. Extract the brief silently from <description>, write native-app-plan.md placeholder, fall through to Step 2c. The user's next interaction is the cost-estimate gate.
3 / 4	one-tap	Skip the multi-select. Extract the brief, show it once, ask only "Look right? (yes / adjust)". On yes → Step 2c. On adjust → fall through to walk-through.
≤ 2 / 4	walk-through	Current behaviour. Run the multi-select feature picker described in Step 2b.1, then the brief confirmation.

Print the chosen tier so the user knows which path is running:

"→ Prompt richness: 4/4 — skipping discovery questions, extracting brief and going straight to the plan-cost preview." (or 3/4 / ≤2/4 with the matching path name)

--full-discovery escape hatch: if $ARGUMENTS contains --full-discovery, force walk-through regardless of score. Use this in dogfood runs where you want to exercise the multi-select path.

--no-discovery escape hatch: if $ARGUMENTS contains --no-discovery, force auto-plan regardless of score. Use this for fully-headless runs from the wrapper templates repo.

Step 2b.1 — Walk-through path (only when tier = walk-through)

Read references/requirements-discovery.md. Infer context-aware options from the user's description, ask exactly one structured AskUserQuestion, and never use markdown checkboxes in the question text.

Wait for the user's response. Summarize their answers into a requirements brief — 4–8 bullet points covering what users can do, what data is tracked, and integrations.

Confirm once:

"Here's the brief I'll use for planning: • (bullet 1) • (bullet 2) ... Look right? (yes / adjust)"

Store the confirmed brief as <requirements_brief>. This replaces the thin $ARGUMENTS as the primary input to the planner.

Step 2b.2 — One-tap path (tier = one-tap)

Skip the multi-select question. Extract a 6–10 bullet brief directly from <description> covering: user roles, key entities, primary workflow, severity / status enums if present, integrations / connectors, native capabilities, and any explicit constraints. Show it with a single confirm:

"Your description is detailed enough to skip the feature picker. Here's the brief I extracted: • (bullet 1) • (bullet 2) ... Look right? (yes / adjust / start over)"

• yes → store as <requirements_brief>, fall through to Step 2c.
• adjust → drop to Step 2b.1 (walk-through) so the user can edit via the multi-select.
• start over → return to Step 2a and re-prompt for the description.

Step 2b.3 — Auto-plan path (tier = auto-plan)

Skip both the multi-select AND the brief confirmation. Extract the brief silently and store it as <requirements_brief>. Print it as a transparency log only:

"→ Auto-plan tier (4/4). Extracted brief from your description: • (bullet 1) • (bullet 2) ... → Going straight to the plan-cost preview (Step 2c). The brief above is locked in unless you abort there."

Do not ask for confirmation here — the user agreed to this when their prompt scored 4/4. The plan-preview gate at Step 2c remains in force as the last cheap exit before any side effects.

Step 2b.4 — Common to all paths

Auto-proceed after yes (or after auto-plan transparency log). Fall through directly to Step 2c (plan preview). Do NOT add a separate "Proceed to planning?" prompt — the brief confirmation IS the planning go-ahead. The only abort gate after this is Step 2c's proceed/edit/abort block, which is intentionally distinct because it shows the rough cost estimate.

Classify Dataverse planning before Step 2c and stash <dataverse_planning_mode>:

• connector-only only when every record source and write target is an explicit non-Dataverse connector/system of record, and the app needs no app-owned persistent rows, Dataverse offline data, retained File/Image artifact, existing Dataverse table, or Dataverse-backed native capability.
• required for every other case, including ambiguity. Do not infer connector-only merely because the brief names a connector.

Also stash <exact_target_facts_required> = yes when planning depends on any existing/standard/managed table, reuse or extension decision, proposed-name collision decision, relationship target, computed column, or target customizability fact. Otherwise set it to no. This flag controls only safe planning degradation; it never relaxes /add-dataverse reconciliation.

Now execute the deferred Step 1.7 publisher-prefix detection only when <dataverse_planning_mode> = required. For connector-only, set $DETECTED_PUBLISHER_PREFIX = "", print ↷ Publisher-prefix discovery skipped — connector-only planning., and do not call detect-publisher-prefix.js.

Design decisions are deferred to Step 6.75 — /design-system (ships with this plugin) handles brand inputs, the style picker, and visual companion preference in one flow after the project is scaffolded. Do NOT ask design questions here.

Set tentative defaults (used by Step 3b before /design-system runs):

• <visual_companion> = yes — open _plan_preview.html in browser at Gate 4 by default. /design-system at Step 6.75 may downgrade this to no (path (d) in its cost picker), persisted to memory-bank for future runs.
• <design_vibe_opt_in> = deferred — Step 6.75 sets the real value. While deferred, the planner does NOT prompt for a direction; it writes a placeholder ## Design Direction: <deferred — set by /design-system> block so screen-planner can still run.

--no-design escape hatch. For headless / token-constrained runs, set --no-design in $ARGUMENTS. It forces <visual_companion> = no, skips the style-picker handoff at Step 3a entirely, and short-circuits Step 6.75 to a no-op (placeholder block stays in native-app-plan.md; screen-builders fall back to industry-inferred defaults).

Step 2c — Plan preview (rough, always shown)

Goal: Give the user a cheap exit before any mutation happens. This is the last point in the flow with zero side effects — no git clone, no npm install, no npx power-apps init, no agent tokens spent on planning. After Step 3 starts, every abort gets more expensive (half-written native-app-plan.md, partial _screens_section.md, architect tokens already burnt).

Always runs. There is no --no-preview flag in v0 — we need calibration data (~10+ runs with recorded estimate-vs-actual) before we can trust the rough estimates enough to let users skip them. Once the data shows estimates are reliably within ±50%, evaluate adding a skip flag for repeat-user workflows.

Compute the estimates from inputs already in hand (no agent spawn — pure heuristics on the confirmed brief and the wizard answers):

Output	Input proxy	Computation	Confidence
Tables	Distinct nouns in confirmed brief	count(unique_nouns) × [0.7, 1.3] rounded	low — architect may merge or split
Connectors	Step 2b inferred connector list	len(inferred) (already exact)	high
Screens	Confirmed features in brief	count(features) × [2, 3]	low — depends on navigation choice
Planning min	Tables + screens	lower bound max(10, tables × 0.3 + screens × 0.4 + 2); upper bound max(15, computed upper)	low — protects the quality-first Gate 1 budget
Scaffold min	Fixed	1-2 (template preparation + npm install already happened before skill invocation)	high
Build min	Screens, parallel cap of 5	ceil(screens / 5) × 0.6	medium
Extra prompts	<industry_confidence> + <design_vibe_opt_in>	+1 if low-confidence industry; +1 if vibe-opt-in == yes	high

Print the block once, exactly in this format (substitute computed values; ranges as low-high):

─── Plan preview (rough) ─────────────────────────────────
Based on your confirmed brief, before any agent runs:

Scope (proxy estimates — actual numbers come from architects):
  Tables       ~<low>-<high>      ← from <N> nouns in brief; architect may merge/split
  Connectors    <N> inferred      ← <comma-separated names>  (confirm at Gate 3)
  Screens     ~<low>-<high>       ← from <N> features × ~2-3 screens each
  Approval gates  4               ← fixed (data model, native, connectors, screen plan)

Time (rough — agent time only, excludes your approval latency at gates):
  Planning      ~<low>-<high> min ← includes the quality-first 10–15 min data-model target; approvals add latency
  Scaffolding   ~1-2 min          ← validates prepared template + runs power-apps init
  Screen build  ~<low>-<high> min ← parallel, capped at 5 concurrent

Token tier: Opus everywhere in v0 (model routing not yet shipped).

⚠ These are proxies, not measurements:
  • Table count is "noun count in brief" — architect may collapse or split
  • Time excludes your approval latency at the 4 gates
  • If industry inference is low-confidence, +1 picker prompt
  • If you opted into the design vibe picker, +1 prompt + planner re-spawn
  • If any gate is rejected, that section regenerates (~2-3 min each)

Proceed, edit brief, or abort? [proceed/edit/abort]
─────────────────────────────────────────────────────────

Three-option exit:

User answer	Action
proceed (or empty / Enter)	Continue to Step 3. Default.
edit	Jump back to Step 2b. Re-confirm the brief with the user's changes. After 2b re-confirms, return here for a fresh preview. No working dir mutations — Step 2c runs before mkdir -p <working_dir> in Step 3.
abort	Print "Aborted at Step 2c. No files created. Re-run /create-mobile-app when ready." and exit cleanly. No working dir, no memory bank, no scaffold.

Why "always show" is correct in v0 (do not skip without explicit user request):

• Cost when user proceeds: ~30s (read + decide). Token cost ~500/run = ~$0.008.
• Cost when user aborts late (after Step 3 starts): 5-10 min + dirty working dir + frustration.
• Asymmetry: bounded 30s vs unbounded 30 min. Always show the bounded cost.
• Forced calibration: every run produces the <estimate, actual> data we need for v0.x model routing decisions. Skipping drops calibration data.

Set expectations before handing off to the planner:

"Brief locked in. Planning surfaces 4 approval prompts (data model → native capabilities → connectors → screens). Data-model readiness is quality-first, with a 10–15 minute target: • Gate 1 (data model) — budget 10–15 min for verified reuse/extend/create decisions, ER columns, relationships, tiers, and risks • Gate 2 (native capabilities) — ~10s (quick) • Gate 3 (connectors) — ~30–60s • Gate 4 (screens + design) — 3–8 minutes (this is the heavy one: design vibe picker if opted in, then per-screen specs and HTML preview generation)

For Dataverse-required apps, factual foreground milestones will show environment, inventory, candidate, detail, and timing counts within 30 seconds. Connector-only apps skip those metadata milestones. While the architect runs, new milestone IDs from .tmp/data-model-planning-status.json are rendered without inventing percentages. If Gate 1 has not surfaced after 15 minutes, inspect the last applicable milestone before interrupting."

Step 2d — Template-only mode

No background scaffold pipeline is used. The template is already present in <working_dir> and dependencies are expected to be installed before this skill starts (npm install). Continue directly to Step 3.

Step 3 — Plan (planner agent + 4 approval gates)

Telemetry checkpoint: planning

First, create the working and planning-artifact directories:

mkdir -p <working_dir> <working_dir>/.tmp

Step 3.0 — Foreground Dataverse planning snapshot and evidence

Planning stays read-only. Branch on <dataverse_planning_mode>:

• connector-only — skip every command in this section. Set SNAPSHOT_PATH and EVIDENCE_PATH to empty/not supplied, print ↷ Foreground planning snapshot skipped — the confirmed brief is connector-only., and continue to planner dispatch. Connector-only planning does not perform Dataverse metadata reads; the skill's existing global prerequisites remain unchanged.
• required — resolve the already selected environment again in the foreground and create one normalized foreground planning snapshot as below. Do not make the nested planner or architect rediscover the tenant.

PLANNING_ENV_JSON=$(node "${PLUGIN_ROOT}/scripts/resolve-environment.js" "$ACTIVE_ENV_ID")
ACTIVE_ENV_URL=$(node -e "const j=JSON.parse(process.argv[1]); console.log(j.environmentUrl || '')" "$PLANNING_ENV_JSON")
ACTIVE_TENANT_ID=$(node -e "const j=JSON.parse(process.argv[1]); console.log(j.tenantId || '')" "$PLANNING_ENV_JSON")
test -n "$ACTIVE_ENV_URL" -a -n "$ACTIVE_TENANT_ID" || {
  echo "✗ Foreground planning snapshot requires a resolved Dataverse URL and tenant."; exit 2;
}
echo "✓ Planning environment resolved: $ACTIVE_ENV_URL (tenant $ACTIVE_TENANT_ID)"

Build <DATAVERSE_CONCEPTS> from every domain noun and workflow family in the approved brief, not a sample. Preserve multiword/header-child families such as medical assessments, care activities, release events, custody transfers, test results, and evidence attachments. Add known standard or required-existing logical names to <EXPLICIT_TABLES>. Build <PROPOSED_TABLES> from the detected publisher prefix for every clearly proposed custom table so collisions and missing names are explicit; leave a name out rather than inventing it when the concept is not yet stable.

Detailed advisory discovery is quality-bounded:

• Keep the complete customizable-table inventory and ranking.
• Required exact-name tables are always detailed and do not consume advisory capacity.
• A concept credibly covered by an exact table does not receive speculative advisory alternatives.
• Every unresolved concept receives its best advisory candidate first.
• Only then allocate second/third candidates, with at most 3 per concept and a target ceiling of 40 unique advisory tables.
• When more than 40 unresolved concepts have distinct best candidates, exceed 40 only enough to preserve one candidate per concept. Quality coverage takes priority over the target ceiling.
• Inventory-only alternatives remain available for the existing one-time bounded exact-name expansion.

SNAPSHOT_PATH="<working_dir>/.tmp/dataverse-foreground-planning-snapshot.json"
EVIDENCE_PATH="<working_dir>/.tmp/dataverse-planning-evidence.md"

node "${PLUGIN_ROOT}/scripts/create-dataverse-snapshot.js" \
  --env-url "$ACTIVE_ENV_URL" \
  --tenant-id "$ACTIVE_TENANT_ID" \
  --output "$SNAPSHOT_PATH" \
  --concepts "<DATAVERSE_CONCEPTS>" \
  --tables "<EXPLICIT_TABLES>" \
  --proposed-tables "<PROPOSED_TABLES>"

node "${PLUGIN_ROOT}/scripts/render-dataverse-planning-evidence.js" \
  --snapshot "$SNAPSHOT_PATH" \
  --output "$EVIDENCE_PATH"

node -e '
  const s=require(process.argv[1]);
  const t=s.timings;
  const d=s.detailLoadSummary;
  console.log(`✓ Dataverse inventory: ${s.inventoryFacts.customizableTables} customizable + ${s.inventoryFacts.exactNameTables} bounded exact-name discoveries (${s.inventoryFacts.requiredExactNameTables} required, ${s.inventoryFacts.proposedCollisionTables} proposed collisions) (${t.inventoryRetrievalMs} ms)`);
  console.log(`✓ Candidate selection: ${s.candidateRanking.length} concepts → ${d.attemptedCandidates} detailed candidates (${d.requiredCandidates || 0} required, ${d.advisoryCandidates || 0} advisory, ${d.exactCoveredConcepts || 0} exact-covered concepts; ${t.candidateSelectionMs} ms)`);
  console.log(`✓ Detail loading: ${d.attemptedCandidates} attempted, ${d.loadedCandidates} loaded, ${d.failedCandidates} failed; ${s.tables.reduce((n,x)=>n+x.facts.columnCount,0)} columns, ${s.tables.reduce((n,x)=>n+x.facts.relationshipCount,0)} relationships, ${s.tables.reduce((n,x)=>n+x.facts.keyCount,0)} keys (${t.detailLoadingMs} ms)`);
  console.log(`✓ Exact names: requested [${s.exactNameResolution.requestedTables.join(", ")}], loaded [${s.exactNameResolution.loadedTables.join(", ")}], unavailable [${s.exactNameResolution.unavailableTables.join(", ")}]`);
  console.log(`✓ Proposed names: ${s.proposedNameChecks.collisions.length} collisions, ${s.proposedNameChecks.missing.length} missing; foreground planning snapshot total ${t.totalDurationMs} ms`);
' "$SNAPSHOT_PATH"
echo "✓ Planning evidence: $EVIDENCE_PATH"

If environment resolution, token acquisition, inventory, required exact-name metadata/detail loading, parsing, or evidence rendering fails, surface the exact failure and do not treat an unreadable response as an empty inventory:

• For every required Dataverse plan, stop planning with a visible BLOCKED: Dataverse planning metadata unavailable for exact target decisions result. Do not dispatch a snapshot-only architect and do not proceed toward Dataverse mutation. The mutation workflow does not accept an unresolved Unverified plan as an executable contract.
• A concept-selected candidate is advisory unless it is also named by --tables. If advisory detail metadata is unsupported, abstract, or inaccessible, keep the snapshot, record it in detailLoadFailures, list it in the evidence appendix, and continue. Explicit --tables and bounded exact-name expansions remain required and fail closed.
• --proposed-tables performs collision checks only. Missing proposed names are not required-table failures and must never be selected for detail loading solely because they were proposed.

The snapshot script emits factual DATAVERSE_SNAPSHOT_PROGRESS lines after inventory, candidate selection, and detail loading. Print them immediately; never replace them with estimates or percentages. Once the architect starts, watch <working_dir>/.tmp/data-model-planning-status.json and print each new milestone ID once with its counts and elapsed time. The first environment or snapshot milestone must be visible within 30 seconds. The foreground orchestrator owns this rendering; the architect only owns the status artifact.

For required, pass SNAPSHOT_PATH and EVIDENCE_PATH verbatim to the planner prompt and every direct data-model-architect fallback/revision. A supplied matching snapshot activates the architect's snapshot-only path: no Bash discovery and no live Dataverse calls inside the agent. For connector-only, pass the mode explicitly and state that both paths are not supplied; never provide placeholder file paths.

Benchmark method and acceptance criteria: references/dataverse-planning-benchmark.md.

Hard rule — planner writes are restricted during Step 3. The planner (and any sub-agents it spawns) is permitted to write to only:

• <working_dir>/native-app-plan.md
• <working_dir>/_screens_section.md
• <working_dir>/.tmp/*

All other paths in <working_dir>/ (notably app/, src/, package.json, power.config.json, tamagui.config.ts, tsconfig.json, node_modules/, memory-bank.md) are owned by the foreground setup phases. Do not mutate them during planning.

If the planner needs to record a DONE_WITH_CONCERNS from a sub-agent (data-model architect, screen-planner), add it to an in-memory queue DEFERRED_CONCERNS[] during Step 3. Do not write memory-bank.md yet. Step 6.7 must always flush DEFERRED_CONCERNS[] into memory-bank.md ## Concerns immediately after the file is created.

Resume-from-draft check. Before spawning, check if <working_dir>/native-app-plan.md already exists with content. If yes, a previous planner run (possibly in a degraded context with no Task/gate tools) already drafted sections. Read it. If it has populated ## Data Model / ## Native Capabilities / ## Connectors but the gates were never run (no ## Approvals block, or the file was authored by an agent that returned BLOCKED: tool surface missing), pass resume_from_draft: true and the existing path to the planner so it loads the draft as baseline instead of regenerating from scratch.

Planner preflight (silent). Before the full Task spawn, do a no-op Task probe for mobile-app:native-app-planner (same pattern as Step 11.0). If the probe fails with Agent type … not found, tool unavailable, or the host clearly cannot route nested agents, fall through to inline-gate mode (described below) without prompting. The orchestrator has the full tool surface itself — it can run the gates directly. Do not retry, do not ask the user.

Announce the handoff before the Task call (so the user isn't staring at a blank screen while the planner spins up):

• required: > "→ Spawning planner agent from the verified foreground planning snapshot. Gate 1/data-model readiness is quality-first with a 10–15 minute target. I will print each factual data-model-planning-status.json milestone and elapsed count as it lands."
• connector-only: > "→ Spawning planner agent in connector-only mode; a foreground planning snapshot and data-model mutation are not required."

Then spawn the mobile-app:native-app-planner agent via Task (the plugin name mobile-app: prefix is required — without it Task returns Agent type not found):

Spawn agent: mobile-app:native-app-planner

Prompt:
  Plan a Power Apps mobile app.

  Requirements brief (confirmed with user):
  <requirements_brief — bullet points from Step 2b>

  Design vibe opt-in: <design_vibe_opt_in — always "deferred" unless `--no-design` is in $ARGUMENTS, in which case use "skip". Never invent yes/no/other values.>
  Visual companion: <visual_companion — "yes" or "no">

  Original prompt: <full $ARGUMENTS verbatim>
  Wizard answers: <Step 2 answers>
  Working directory: <absolute path of <working_dir>>
  Plugin root: ${PLUGIN_ROOT}
  Dataverse planning mode: <required | connector-only>
  Dataverse planning failure reason: none
  Normalized Dataverse foreground planning snapshot: <absolute SNAPSHOT_PATH verbatim for required; otherwise NOT SUPPLIED>
  Dataverse planning evidence: <absolute EVIDENCE_PATH verbatim for required; otherwise NOT SUPPLIED>
  Structured schema contract: <absolute
  `<working_dir>/.tmp/dataverse-schema-contract.json` for required; otherwise
  NOT SUPPLIED>
  Publisher prefix (detected from env): <DETECTED_PUBLISHER_PREFIX from Step 1.7, e.g. "cr8142a" — use literally as `<prefix>_<entity>` in all logical names. If empty/NOT DETECTED, fall back to `cr` placeholder and surface a `DONE_WITH_CONCERNS` note that Dataverse will normalize at create time.>

  Follow native-app-planner.md. Run all 4 approval gates. On terminal return, emit one of `DONE` / `DONE_WITH_CONCERNS:` / `NEEDS_CONTEXT:` / `BLOCKED:` as the literal first line per AGENTS.md rule #10.

The planner runs gates internally for data model → native capabilities → connectors → screen plan, and writes <working_dir>/native-app-plan.md. Wait for it to return before continuing — do not proceed on a partially-approved plan. On a successful required return, require both .tmp/dataverse-schema-contract.json and .tmp/mobile-plan-status.json before continuing. If the receipt is missing, STOP as BLOCKED; this orchestrator must not synthesize it after the planner has returned.

3.0a — Inline-gate fallback (planner unavailable OR returned BLOCKED: tool surface missing)

When the preflight fails OR the planner returns BLOCKED: tool surface missing <…>, the orchestrator runs the four gates inline. Do NOT re-spawn the planner — it cannot succeed in this host. Print once:

"→ Planner agent unavailable in this host — running approval gates inline. (No action needed; this is automatic.)"

Then execute, in order, using your own EnterPlanMode + AskUserQuestion:

1. If a draft native-app-plan.md exists: read it as baseline. Surface each populated section (## Data Model, ## Native Capabilities, ## Connectors) one at a time via EnterPlanMode, take user feedback inline, edit the file in place. Skip generating sections that are already populated and approved.

2. If no draft exists: spawn mobile-app:data-model-architect directly via Task (single architect, not the orchestrator agent) to draft ## Data Model; then build ## Native Capabilities + ## Connectors inline from the brief; then spawn mobile-app:screen-planner with phase: graph and phase: specs per the two-phase Gate 4 split.

   Before each screen-planner spawn, print a one-line ETA so the user knows the agent is live and roughly how long to wait (the agent's own Bash echo progress markers — see agents/screen-planner.md "Progress streaming" — surface every milestone, but the orchestrator's pre-spawn line gives the wall-clock budget):

   • Before phase: graph: > "→ [Gate 4a] Spawning screen-planner phase=graph (~2 min for ${N} screens)…"
   • Before phase: specs: > "→ [Gate 4b] Spawning screen-planner phase=specs (~1 min/screen, ~${N} min for ${N} screens). Progress markers will appear inline."

MUST forward the Dataverse planning mode in the direct architect prompt. In required, also forward SNAPSHOT_PATH and EVIDENCE_PATH verbatim and do not resolve the environment or run Dataverse discovery again. In connector-only, state that both paths are not supplied; never invent placeholder artifacts.

MUST forward $DETECTED_PUBLISHER_PREFIX from Step 1.7 in the architect prompt — same line as the planner prompt at Step 3 line 1034: "Publisher prefix (detected from env): <DETECTED_PUBLISHER_PREFIX> — use literally as <prefix>_<entity> in all logical names. If empty/NOT DETECTED, fall back to cr placeholder and surface a DONE_WITH_CONCERNS note that Dataverse will normalize at create time." Without this, the architect defaults to cr_ and the whole plan needs a post-hoc sweep when the real prefix is something else (e.g. cr3e9).

In required, also require the direct architect to write and normalize <working_dir>/.tmp/dataverse-schema-contract.json per its agent contract. A draft Markdown section without that sidecar is not an executable Gate 1 result.

Why this works even though the planner just returned BLOCKED for tool surface: the orchestrator (this skill, running in the user's slash-command session) always has the full tool surface — Task, EnterPlanMode, ExitPlanMode, AskUserQuestion, Read, Write, Bash. What's missing is the surface inside nested agent contexts (the native-app-planner agent runs in a sandbox without EnterPlanMode/AskUserQuestion, which is why its Step 0 preflight returned BLOCKED). The leaf agents data-model-architect and screen-planner only need Read/Write/Bash to draft markdown — they don't need EnterPlanMode/AskUserQuestion themselves. Spawn them; the orchestrator owns the gates.

3. Run the gates yourself — use EnterPlanMode four times (data model → native caps + connectors merged → screen graph 4a → screen specs 4b). Same gate prompts as the planner agent would use. Gate 4 is a markdown screen-graph review only — design picking happens unconditionally at Step 6.75 via /design-system (no separate style-picker handoff at Gate 4 even in inline mode).

4. Write the final approved native-app-plan.md with an ## Approvals block at the bottom listing each gate, who approved (user), and a timestamp.

   HARD RULES for the plan structure (mirror the planner agent's template at agents/native-app-planner.md Step 4):

   • Top-level headings are EXACTLY: ## Overview, ## App Requirements, ## Data Model, ## Native Capabilities, ## Design Direction, ## Connectors, ## Screens, ## Approvals. Do NOT invent a ## Brief super-section that nests the data model under it.
   • ## App Requirements is the user's confirmed brief verbatim (the <requirements_brief> from Step 2b), capped at ~80 lines. No expansion, no rewriting, no embedded preview of the data model.
   • Discovery failure notes (e.g. az login on the wrong tenant, 401 from dataverse-request.js, all entities classified Create) go to <working_dir>/memory-bank.md under ## Discovery Notes, NOT into the plan. Keep at most a single one-line breadcrumb in ## Data Model like > Discovery skipped — see memory-bank.md. if relevant.
   • Sample data notes, immutability plug-in notes, file-column setup notes, dispatch-block server rules go under a single ### Notes subsection in ## Data Model. Cap each at 2 sentences; link to post-deployment-tasks.md for longer write-ups instead of inlining.

5. Record the same structured approval receipt as the planner path. At data-model acceptance, initialize <working_dir>/.tmp/mobile-plan-status.json with the exact normalized contract content/hash. After each later gate is accepted, update only that gate's approval record and the current plan hash; after Gate 4b, record the final structured service dependencies and integrity hash. Follow agents/native-app-planner.md Step 6 exactly. Never call the operation manifest builder to create or restamp this receipt. A changed approved section invalidates its record until the existing inline gate approves it again.

If the orchestrator's OWN Task tool is unavailable (rare — would mean even leaf agents can't be spawned), fall further to fully-inline mode. In required, draft the data model from SNAPSHOT_PATH plus EVIDENCE_PATH with no live OData probe and write/normalize the same structured schema contract required by agents/data-model-architect.md. In connector-only, write an explicit zero-table/no-Dataverse ## Data Model section and no contract. Then draft native caps + connectors heuristically, draft the screen graph + specs against shared/references/screen-templates.md, and run the four gates against the user. This is the last-resort path — functional but slower because the orchestrator does work the architects normally parallelize.

Hard rule: never silently skip a gate just because the planner couldn't run. The user MUST approve each section through EnterPlanMode before any mutation step (Step 8 onwards) executes.

3.0 — Sub-agent return-status switch (canonical)

Use the plugin-wide protocol in AGENTS.md rule #10 for every Task return in this skill: planner, parallel screen-builders, and future agent spawns. Parse the literal first line and branch: DONE continues; DONE_WITH_CONCERNS: surfaces + records in memory-bank.md; NEEDS_CONTEXT: re-dispatches with missing context, capped at 2 retries; BLOCKED: stops and records under ## Blocks. Unknown first lines are malformed and must be treated as BLOCKED.

Data-model exact-name expansion: when the planner or direct architect returns exactly NEEDS_CONTEXT: detailed-dataverse-metadata:<logical names>, sort and de-duplicate those names. This signal is valid only in required mode with a validated base snapshot, whether it came from the planner or the direct architect fallback; receiving it in connector-only mode is BLOCKED. Perform one bounded foreground expansion. Reuse the existing snapshot inventory, issue at most one exact-name metadata query for requested names absent from it, and do not run another broad inventory query:

node "${PLUGIN_ROOT}/scripts/create-dataverse-snapshot.js" \
  --env-url "$ACTIVE_ENV_URL" \
  --tenant-id "$ACTIVE_TENANT_ID" \
  --base-snapshot "$SNAPSHOT_PATH" \
  --output "$SNAPSHOT_PATH" \
  --tables "<exact comma-separated logical names>"

node "${PLUGIN_ROOT}/scripts/render-dataverse-planning-evidence.js" \
  --snapshot "$SNAPSHOT_PATH" \
  --output "$EVIDENCE_PATH"

node -e '
  const s=require(process.argv[1]);
  const x=s.expansion;
  const d=s.detailLoadSummary;
  console.log(`✓ Expansion requested: [${x.requestedTables.join(", ")}]`);
  console.log(`✓ Expansion loaded: [${x.loadedTables.join(", ")}]`);
  console.log(`✓ Expansion unavailable: [${x.unavailableTables.join(", ")}]`);
  console.log(`✓ Expansion details: ${d.attemptedCandidates} attempted, ${d.loadedCandidates} loaded, ${d.failedCandidates} failed`);
  console.log(`✓ Expansion timing: metadata ${s.timings.inventoryRetrievalMs} ms, selection ${s.timings.candidateSelectionMs} ms, details ${s.timings.detailLoadingMs} ms, total ${s.timings.totalDurationMs} ms`);
' "$SNAPSHOT_PATH"

Print the expansion's requested/loaded/unavailable names and timings immediately, then re-dispatch the same planner or architect once with the same snapshot/evidence paths. A second detailed-metadata signal is BLOCKED; do not loop, broaden concepts, or defer exact validation to mutation.

Data-model proposed-name expansion: when the planner or direct architect returns exactly NEEDS_CONTEXT: proposed-dataverse-names:<logical names>, sort and de-duplicate those names. This signal is valid only in required mode with a validated snapshot. Perform one collision-only foreground expansion:

node "${PLUGIN_ROOT}/scripts/create-dataverse-snapshot.js" \
  --env-url "$ACTIVE_ENV_URL" \
  --tenant-id "$ACTIVE_TENANT_ID" \
  --base-snapshot "$SNAPSHOT_PATH" \
  --output "$SNAPSHOT_PATH" \
  --proposed-tables "<exact comma-separated logical names>"

node "${PLUGIN_ROOT}/scripts/render-dataverse-planning-evidence.js" \
  --snapshot "$SNAPSHOT_PATH" \
  --output "$EVIDENCE_PATH"

This expansion checks collisions only; it does not treat absent proposed names as required existing tables or load their details. Re-dispatch once. A second proposed-name signal is BLOCKED. If a collision is found and the architect needs compatibility facts, it may then use the separate one-time detailed-dataverse-metadata expansion for that existing table.

Planner-only early-return signals are handled before the status switch: INDUSTRY_CONFIRM_REQUESTED: routes to Step 3.0a; DESIGN_VIBE_REQUESTED: routes to Step 3a. After the handoff, re-spawn the planner and process its new first line through this switch.

Step 3.0a — Industry confirmation handoff (orchestrator-owned)

When the planner is uncertain about which industry the app belongs to (no keyword match, ambiguous match, or wizard-aesthetic conflict), it returns early with this single line as its message:

INDUSTRY_CONFIRM_REQUESTED: <inferred-industry>|<reason-code>|<top-3-alternatives-comma-sep>

Example: INDUSTRY_CONFIRM_REQUESTED: productivity|no-keywords|field-ops,healthcare,e-commerce

This fires before Gate 1 — it's not a gate, just a confidence check so the wrong industry doesn't silently lock in the design language for the entire app.

Skip this section if <design_vibe_opt_in> is yes or skip — in those cases the user is either driving design explicitly (yes) or has opted out of design entirely (skip), so industry inference doesn't matter.

When you see INDUSTRY_CONFIRM_REQUESTED: and <design_vibe_opt_in> is no:

1. Parse the three pipe-delimited fields. Map reason codes to a short user-facing explanation:

   • no-keywords → "no clear industry signal in your description"
   • ambiguous-match → "your description matches multiple industries"
   • wizard-conflict → "your aesthetic answer doesn't match the inferred industry"

2. Map each industry slug to a one-line description for the picker (use these exactly):

   Slug	Description
   field-ops	Field/Ops — high contrast, large targets, camera-forward (Uber Driver, ServiceTitan)
   finance	Finance — blue palette, conservative type, generous whitespace (banking apps)
   healthcare	Healthcare — warm palette, friendly type, compassionate copy (patient apps)
   education	Education — bright playful, gamification, streak/progress (Duolingo)
   productivity	Productivity — near-monochrome, dense layout, monospace data (Linear, Notion)
   e-commerce	E-commerce — brand-forward color, product imagery, frictionless CTAs (retail apps)
   tech-iot	Tech/IoT — dark + accent gradients, data-dense cards, real-time indicators (monitoring dashboards)

3. Ask one AskUserQuestion:

   "Quick sanity check before I build the design: I inferred this is a app, but . Confirm or pick another:

   (a) — recommended (b) (c) (d) (e) Other / let me describe — free text

   Which? (a / b / c / d / e — default: a)"

4. Persist the answer:

   echo "<chosen-industry-slug>" > "<working_dir>/.industry-confirmed"
   

   For option (e), let the user free-text a description; map it to the closest slug (or productivity as final fallback) and store that.

5. Re-spawn the planner. Use the same prompt as Step 3, plus an extra line:

   Industry confirmed: <chosen-industry-slug>
   

   The planner will see this on re-spawn, skip its detection + confidence check, and lock the industry to your value. After the re-spawn, re-check the planner's return value — it may now return normally, or it may still return DESIGN_VIBE_REQUESTED: (handled in Step 3a) if the user opted into the vibe picker.

Step 3a — Style-picker handoff (no-op in current plugin layout)

/design-system ships with this plugin and always runs at Step 6.75, so the style-picker handoff at Gate 4 is a no-op. Behavior:

• The planner writes a placeholder ## Design Direction: <deferred — set by /design-system> block into native-app-plan.md at Gate 4 and proceeds without asking the user. Step 6.75 rewrites the placeholder with the real direction.
• If a legacy planner output emits DESIGN_VIBE_REQUESTED: as its first line, write the placeholder block yourself (insert before ## Design, or before ## Screens if ## Design is absent), then re-spawn the planner with Design vibe opt-in: done. Do NOT run a vibe picker here — Step 6.75 owns that.
• If --no-design is in $ARGUMENTS, write the placeholder block, mark <design_vibe_opt_in> = skip, and Step 6.75 also no-ops. Screen-builders fall back to industry-inferred defaults from universal-patterns.md.

If the planner's first return is anything other than DESIGN_VIBE_REQUESTED: — i.e. it ran all gates including Gate 4 normally — skip directly to Step 3b.

Step 3b — Open the plan preview in the user's browser (orchestrator-owned)

The planner emits a line of the form PLAN_PREVIEW_PATH: file://<abs-path>/_plan_preview.html before each Gate 4 plan-mode entry. The planner itself does NOT open the browser — sub-agent shells often lose GUI context, and silent open-failures leave the user staring at the spinner with no preview. The orchestrator owns this step because it has the user's interactive session.

When to run this: every time the planner enters or re-enters Gate 4 (initial pass + each reject loop). Detection: scan the planner's most recent visible output for the PLAN_PREVIEW_PATH: token; the value after the colon is the absolute file:// URL.

What to do:

1. Print the link in a dedicated message so the user always has the fallback (clickable in most terminals):

   "Plan-time visual preview: file:///_plan_preview.html"

2. If <visual_companion> = no, stop here. Do not attempt to open a browser. The user explicitly opted out; the printed link is their handle. Continue immediately to the planner's Gate 4 prompt.

3. Else attempt to open in the user's default browser via the OS-portable chain:

   open "<abs-path>/_plan_preview.html" 2>/dev/null \
     || xdg-open "<abs-path>/_plan_preview.html" 2>/dev/null \
     || powershell.exe -NoProfile -Command "Start-Process '<abs-path>\_plan_preview.html'" 2>/dev/null \
     || echo "Auto-open failed. Use the link above."
   

4. Do NOT block on success. If the chain prints "Auto-open failed", the link from step 1 is the user's fallback. Continue immediately so the planner's plan-mode prompt surfaces without delay.

If the planner returns without emitting a PLAN_PREVIEW_PATH: line, that is expected — the planner passes skip_preview: true to screen-planner since /design-system (always installed) renders the single visual preview at Step 6.75 after brand locks. Print:

"→ Gate 4 reviewed structurally. Visual preview will appear at Step 6.75 after /design-system locks your brand tokens (~5 min from now after scaffold)."

…and continue without attempting any browser open. Do not warn or treat this as an error — it is the documented behavior.

3.9 — Post-plan publisher-prefix gate

Before continuing to Step 4, verify the written native-app-plan.md actually uses $DETECTED_PUBLISHER_PREFIX from Step 1.7. Catches both the inline-fallback path missing the prefix and an architect that ignored the instruction.

if [ -n "$DETECTED_PUBLISHER_PREFIX" ]; then
  WRONG=$(grep -oE 'cr[a-z0-9]*_[a-z][a-z0-9_]*' "$WORKING_DIR/native-app-plan.md" \
    | grep -vE "^${DETECTED_PUBLISHER_PREFIX}_" | sort -u || true)
  if [ -n "$WRONG" ]; then
    echo "PLAN PREFIX MISMATCH — expected ${DETECTED_PUBLISHER_PREFIX}_, found:"
    echo "$WRONG"
  fi
fi

If mismatches are reported, sweep native-app-plan.md (and any auxiliary files like .datamodel-manifest.json if already written) replacing the wrong prefix with ${DETECTED_PUBLISHER_PREFIX}_ before Step 4. Do NOT proceed to Step 5 with a wrong-prefix plan — the sweep cost grows ~500 occurrences once services are generated.

For required, apply the same prefix correction to .tmp/dataverse-schema-contract.json, then normalize it. Before Step 4, require both approved artifacts:

test -f "$WORKING_DIR/native-app-plan.md"
test -f "$WORKING_DIR/.tmp/dataverse-schema-contract.json"
node "${PLUGIN_ROOT}/scripts/build-dataverse-operation-manifest.js" \
  --normalize-contract "$WORKING_DIR/.tmp/dataverse-schema-contract.json" \
  --output "$WORKING_DIR/.tmp/dataverse-schema-contract.json"

Do not fall back to parsing the Markdown ER diagram when the sidecar is missing or malformed; route through the existing planner/direct-architect revision path.

Step 4 — Auth & environment selection

node "${PLUGIN_ROOT}/scripts/resolve-environment.js" "$ACTIVE_ENV_ID"

If the resolved environment doesn't match what the planner used in Step 3, ask the user for the intended environment ID and re-run resolve-environment.js. Capture the environment ID for Step 6.

Step 5 — Prepare existing template

Telemetry checkpoint: scaffold

This step is template-only and foreground-only. Do not clone/copy templates, do not run background scaffold jobs, and do not use any legacy fallback path.

Print before starting:

"→ [Step 5/13] Preparing existing Expo standalone template in <working_dir> …"

Required checks:

cd <working_dir>
test -f package.json && test -f app.config.js && test -f auth.config.json && test -f tamagui.config.ts
test -d node_modules/expo

If any required template file is missing, STOP:

"This folder is not a fresh expo-app-standalone template. Materialize a fresh template with degit into a new folder, run npm install, then rerun /create-mobile-app --working-dir <fresh-template-dir>."

If node_modules/expo is missing, STOP:

"Dependencies are not installed. Run npm install in the template folder, then rerun /create-mobile-app --working-dir <fresh-template-dir>."

If already-created markers appear (memory-bank.md, .datamodel-manifest.json, or src/generated/services/*.ts) and Step 0 did not enter the resume path, STOP. native-app-plan.md is expected here because Step 3 writes the approved plan before template preparation:

"This folder already looks like a created app. For a new app, materialize a fresh expo-app-standalone template with degit into a new folder and rerun this skill there."

Run the deterministic preparation script once:

PREPARE_SCRIPT="${PLUGIN_ROOT}/scripts/prepare-mobile-template.js"
node - "$PREPARE_SCRIPT" <<'NODE'
const { prepareMobileTemplate } = require(process.argv[2]);

// Replace these placeholders with JSON.stringify(...) output so user-provided
// quotes and dollar signs remain data rather than becoming shell syntax.
const result = prepareMobileTemplate({
  workingDir: <JSON_STRING_OF_WORKING_DIR>,
  displayName: <JSON_STRING_OF_DISPLAY_NAME>,
  slug: <JSON_STRING_OF_SLUG>,
});
process.stdout.write(`${JSON.stringify(result)}\n`);
NODE

The script is the only owner of Step 5 mutations. It updates identity, removes only recognized legacy example hooks/query-client files, copies shared helpers only when missing, verifies that TypeScript inherits the host configuration, and structurally verifies the root provider/theme/safe-area contract. It preserves custom navigation, existing helper bytes, offlineProfile, provider props, and the template's @ts-ignore generation boundaries.

Generated ownership boundary: Step 5 must not create, reset, delete, or write anything under src/generated/. Only Power Apps schema/data-source generation commands own that directory. A generated file required later is created by its owning command, never by a placeholder barrel.

The script fails visibly for unsupported root-layout shapes or dangling legacy imports. Do not fall back to a full-file rewrite or regex patch. After it returns successfully, continue to Step 6.

Fix 1 — App identity in app.config.js and package.json

Substitute the hardcoded template values with wizard answers from Step 2:

Find	Replace with
`const APP_NAME = process.env.APP_DISPLAY_NAME
`const APP_SLUG = process.env.APP_SLUG
"name": "powerapps-standalone-app"	"name": "<slug>"

Bundle ID and scheme are left as template defaults — they are fixed across all dev builds and patched by the wrap pipeline at release time.

Fix 2 — Remove only an empty placeholder power.config.json

The preparation script parses power.config.json and removes it only when environmentId is empty or missing. A populated file is preserved and later validated against the approved environment. Do not use an unconditional delete.

Fix 3 — Remove recognized legacy examples without touching generated code

Newer snapshots do not ship the Contacts / Accounts / UserProfile example hooks or the old app-owned query client. The preparation script removes only those recognized files when present. It never traverses or mutates src/generated/; generated models, services, schemas, and barrels remain owned by Power Apps generation commands.

Do NOT overwrite app/(app)/home.tsx here. The current template ships a safe-area-aware, semantic-token starter route. The screen-builder replaces it only when the approved Screen Map assigns that route.

Keep src/hooks/ itself — screen-builders write new hooks into it.

Fix 3b — Scan for dangling imports referencing deleted files (back-compat only)

The preparation script scans app/ and non-generated src/ files after cleanup. Any remaining legacy example import is an explicit failure. Do not replace whole screens or layouts to make the scan pass; use a supported fresh template or repair the precise stale import before continuing.

Fix 6 — Schema generation boundary

app/_layout.tsx imports schemaMap from src/generated/connectorSchemas.ts, which is generated by npm run generate-schemas (the generate-connector-schemas binary from the @microsoft/power-apps-cli devDep). Do not generate an empty schema map during initial scaffold: the template's @ts-ignore boundary lets tsc validate the scaffold without that artifact, and schema generation is more useful after a data source exists or immediately before dev/build entry points.

Do NOT hand-write a stub connectorSchemas.ts — the generated output has a specific shape that downstream code depends on; a placeholder will break npx power-apps push.

Why tsc already passes post-clone (current template, PR #30): the template's app/_layout.tsx and src/playerConfig.ts carry // @ts-ignore comments above the power.config.json and connectorSchemas imports specifically so the project type-checks before power.config.json and connectorSchemas.ts exist. Never strip these @ts-ignore lines — Fix 8 below preserves them when patching app/_layout.tsx to thread the project's tamaguiConfig into PowerAppsProvider, and any future Edit to either file MUST keep them. Removing them resurfaces a tsc failure against missing generated files.

Fix 7 — Seed shared code only when missing

The preparation script creates the shared source directories and copies each approved sample helper only when its destination does not exist. Existing helpers are byte-for-byte preserved, so reruns cannot overwrite user or builder changes.

Fix 8 — Thread the project's tamaguiConfig and active theme into the host provider

The template ships PowerAppsProvider with host-owned light and dark theme defaults. Fix 8 ensures the project tamaguiConfig and color-scheme-driven defaultTheme are present. It preserves explicit theme and darkTheme overrides when an app already has them, but does not add redundant overrides to a fresh template. Step 9b adds explicit themes only when applying generated brand tokens. Do NOT add an outer <TamaguiProvider> — PowerAppsProvider composes it internally and duplicating triggers "useTheme must be used within a TamaguiProvider" warnings on hot reload.

The preparation script edits the existing root layout structurally. It adds missing imports and provider props, then wraps PowerAppsProvider with SafeAreaProvider only when needed. It does not replace the file and it does not wrap <Slot /> with SafeAreaView; each rendered route owns its content edges to avoid double insets.

Key points:

• Do NOT remove the two // @ts-ignore lines. They keep tsc green pre-npx power-apps init.
• Do NOT add an outer <TamaguiProvider> — PowerAppsProvider composes it internally.
• SafeAreaProvider wraps the tree so child screens can call useSafeAreaInsets() without a context error. Each route must use SafeAreaView or explicit insets for its own visible edges.
• tamaguiConfig is imported from '../tamagui.config' (the default export of tamagui.config.ts at project root).
• defaultTheme flips between light/dark via useColorScheme(). /design-system --add-dark-mode later wires per-token dark variants.

Fix 4 — Shared TypeScript configuration

The current template extends @microsoft/power-apps-native-host/config/tsconfig. That host configuration owns the runtime package paths and the six shared-code aliases: @/components, @/hooks, @/utils, @/tokens, @/generated, and @/native. The preparation script verifies this inheritance and does not create a second template-local alias map. Expo Metro consumes the resulting effective TypeScript paths, so no Babel alias plug-in is required.

<Gradient> (used by components/index.tsx) requires expo-linear-gradient. Assume the upstream template ships it — do NOT edit package.json to add it. If npm install (Step 6.5) later reveals the dep is missing, STOP and ask the user to wait for the next template release; do not work around by adding the dep here (same lockdown rule as /add-native).

Do not run npm install inside Step 5 — in template-only mode dependencies must already be installed before the skill starts.

Install note (current template): The template does not read power.config.json during npm install. The Step 6 → Step 6.5 ordering is kept for predictable checkpoints, but do not run npm run generate-schemas during initial scaffold.

Step 6 — Initialize

Print before starting:

"→ [Step 6/13] Running npx power-apps init -t MobileApp to write power.config.json for environment . ~15–30 seconds."

cd <working_dir>
npx power-apps init -t MobileApp --display-name "<displayName>" --environment-id "<environment-id>" --non-interactive

Substitute the approved Step 2 display name and Step 4 environment ID using shell-safe quoting; do not ask for either value again. Step 5 must leave power.config.json absent. If a populated file remains, STOP and report its environment instead of overwriting it or running init again.

Verify power.config.json exists and both its environmentId and appDisplayName match the approved Step 2/Step 4 values. If initialization fails, report the exact error and STOP.

Step 6.5 — Verify dependencies

This step verifies dependencies only. The user must have run npm install before invoking the skill.

[ -d "<working_dir>/node_modules/expo" ] && echo "✓ node_modules present" || echo "✗ missing — run npm install in the template folder and rerun"

If node_modules/expo is missing, STOP. Tell the user to run npm install in the template folder. Do not provision ADO tokens or run npm install from this skill.

Step 6.5b — Root runtime contract verification

Step 5 already performs the idempotent structural update and postcondition checks. Do not mutate _layout.tsx again here. Verify only that the prepared layout still contains SafeAreaProvider, tamaguiConfig, offlineProfile, and the color-scheme-driven defaultTheme. If brand-token wiring has already run, also verify its explicit theme and darkTheme props. If any applicable contract element is missing, rerun the Step 5 preparation script and stop if it reports an unsupported layout.

Step 6.6 — Scaffold TypeScript gate

Print before starting:

"→ [Step 6.6/13] Running scaffold tsc smoke check (~10–30 seconds)."

With node_modules/ populated, run the scaffold TypeScript gate. Do not run npm run generate-schemas here just to produce an empty connectorSchemas.ts; the template is intentionally type-checkable before that file exists, and the script is already run after data-source changes and again before Step 12 starts the dev server.

npx tsc --noEmit

tsc must pass here. If it doesn't, the post-clone surgery in Step 5 (Fixes 1–7) is incomplete — do not proceed to data sources or screen builders. Re-read the Step 5 fixes against the current working dir contents and reapply any missed edit.

This is the Scaffold gate from the TypeScript Gate Policy. If it fails, capture the full error list once, batch-fix scaffold/template causes, and rerun this gate. Do not continue to Step 6.7 or any app-specific mutation until this gate is clean. If the only failure is a missing generated schema import, preserve the template @ts-ignore boundary rather than generating an empty schema artifact.

Step 6.7 — Seed the memory bank

cp "${PLUGIN_ROOT}/shared/memory-bank.md" "<working_dir>/memory-bank.md"

Fill in the Project facts and Power Platform context sections from Steps 2 and 4. From here on, every step appends to the relevant section of <working_dir>/memory-bank.md immediately after success — not at the end. This is what enables Step 0's resume on a future run.

Immediately after creating memory-bank.md, flush any queued planner concerns from DEFERRED_CONCERNS[] into ## Concerns (append-only). This flush is unconditional: if the queue is non-empty, write it now before continuing to Step 6.75.

Also persist the Visual Companion preference so re-runs (/edit-app, /preview-screens, future /design-system runs) honor it without re-asking. Append to the Project facts section:

visual_companion: <yes|no>   # set in Step 2b — controls whether browser previews open automatically

/preview-screens reads this flag when invoked from inside this project; if no, it prints the file path instead of opening. /edit-app reads it to decide whether to re-open _plan_preview.html after a re-plan. The flag is per-project and does not leak across apps.

Step 6.75 — Design system

Print before starting:

"→ [Step 6.75/13] Locking your design system — source of truth for every screen built next. Takes 5 sec to 3 min depending on path."

Skip this step if --no-design is in $ARGUMENTS — placeholder ## Design Direction: <deferred> block stays in the plan, screen-builders fall back to industry-inferred defaults from universal-patterns.md.

Otherwise, invoke /design-system (ships with this plugin):

Invoke skill: /design-system

Arguments:
  --working-dir <working_dir>

The skill detects orchestrator mode (CODE_APPS_NATIVE_ORCHESTRATING=1), collects brand inputs, presents the cost picker (a/b/c/d), runs the internal style picker, writes brand/design-system.md + brand/tokens.ts, renders brand/design-system.html, and returns with status.

Handle the return per the status protocol (AGENTS.md rule #10):

• DONE → continue to Step 7. Record brand_path, tokens_path, direction in memory-bank.
• DONE_WITH_CONCERNS → surface concerns, ask user, continue.
• NEEDS_CONTEXT → surface question, re-invoke with answer.
• BLOCKED → surface error, STOP.

If the user picked path (c) Skip in the cost picker, the skill returns immediately with DONE and no brand/ files. Screen-builders fall back to ## Design Direction — same as today's behavior. But the user still needs a visual preview before code is written — fall through to the "Skip path preview" block below.

After /design-system returns DONE — two branches:

Branch A — brand/ files exist (user picked path a, b, or d)

This is the FIRST and ONLY HTML preview the user sees in the new flow — Gate 4 was a structural-only review (markdown screen-graph, no HTML). /design-system owns rendering of _plan_preview.html at its Sub-step 6.5 using the locked brand tokens. No re-spawn from the orchestrator is needed; the preview is fresh when the skill returns.

Branch B — Skip path preview (user picked path c — no brand/ files)

The user skipped the design system but still deserves to see their screens before code is written. Render a preview with Field/Ops defaults:

1. Print:

   "→ Design system skipped — rendering screen preview with Field/Ops defaults so you can validate the layout before code is written."

2. Render _plan_preview.html — read the screen specs from native-app-plan.md ## Screens section and render key screens (one List + one Form + one Detail, first match per archetype) using the tamagui-html-mapping.md reference and industry-inferred defaults from ## Design Direction. Write to <working_dir>/_plan_preview.html.

3. Open in browser (if <visual_companion> = yes):

   open "<working_dir>/_plan_preview.html" 2>/dev/null \
     || xdg-open "<working_dir>/_plan_preview.html" 2>/dev/null \
     || powershell.exe -NoProfile -Command "Start-Process '<working_dir>\_plan_preview.html'" 2>/dev/null \
     || true
   

4. Auto-continue — no prompt. The user already approved Gates 1–3 via plan-mode and just looked at the preview. A fourth confirmation here adds friction without adding decision power. Print one line and proceed:

→ Preview rendered with default styling. Continuing to Step 7. (Interrupt and re-run /design-system or /edit-app to revise.)

This ensures every path through the flow gets at least one visual preview before screen-builders write code.

Why this matters: under the OLD two-preview flow, the user saw screens at Gate 4 with default Tamagui colors, mentally committed, then the brand re-rendered later — confusing visual whiplash plus ~3–5 min of wasted token spend on the Gate 4 HTML. Under the NEW flow, Gate 4 is a markdown screen-graph (structural only), and the user only ever sees one HTML preview — at Step 6.75, with the locked brand applied. Single visual decision point, no waste.

Step 6.85 — Offline profile (always asked)

Print before starting:

"→ [Step 6.85/13] Asking whether to set up an offline profile…"

Mobile Offline Profiles let users continue working when their device is disconnected — Dataverse syncs queued changes when connectivity returns. This question must be asked on every run of /create-mobile-app — do NOT skip it based on the initial prompt's wording, the feature picker checkboxes, the requirements brief content, or any keyword heuristic. The user's intent for offline support is not reliably inferrable from any earlier step; the only authoritative signal is an explicit answer here. The plugin owns the setup via /setup-offline-profile.

Pre-conditions — only TWO branches skip this question, both with the explicit print line shown:

Condition	Action
App uses zero Dataverse tables (connectors-only — no .datamodel-manifest.json)	Skip — offline profiles are Dataverse-only. Print: ↷ Step 6.85 skipped — no Dataverse tables in this app. Continue to Step 7.
memory-bank.md ## Offline profile already has status: done OR status: not-applicable from a prior run	Skip — answered previously. Print: ↷ Step 6.85 skipped — offline profile already <done|not-applicable> from a prior run. Continue to Step 7.
Anything else (default — including "user didn't say 'offline' anywhere")	Ask the question below.

Important — do NOT add other skip conditions. Earlier versions of this skill had an "auto-detect / keyword scan" preamble (offline · field · remote · ...) that varied the question default based on whether the initial prompt mentioned offline-indicator keywords. That heuristic was removed because it caused the orchestrator to skip the question entirely when no keywords matched, leaving users with a stale or missing profile on the next sync. The question is now unconditional for any Dataverse-backed app that hasn't already answered.

Ask one AskUserQuestion — phrased neutrally, no implication that the user "should" answer Yes or No based on app type:

Question header: Offline support

Question body: "Mobile Offline Profiles let users continue working when their device is disconnected — Dataverse syncs queued changes when connectivity returns. Set one up now?"

Options (default = Yes):

• Yes — create profile now (recommended) — invokes /setup-offline-profile, walks the 3-gate flow, then continues to Step 7
• Skip — I'll add it later via /setup-offline-profile — continues to Step 7 with no profile (memory-bank status stays unset so the question fires again on the next /create-mobile-app run against this dir)
• Skip — this app doesn't need offline support — continues to Step 7; sets memory-bank.md ## Offline profile status: not-applicable so future re-runs don't re-ask

If the user picks Yes, invoke /setup-offline-profile as a sub-skill. It reads the same .datamodel-manifest.json you just wrote (Step 6 /add-dataverse's Step 6d output) — no need to pass arguments. The sub-skill returns DONE / DONE_WITH_CONCERNS: / BLOCKED: per the agent return-status protocol. On BLOCKED:, propagate up. On DONE_WITH_CONCERNS:, surface concerns to the user inline, then continue.

State transfer: /setup-offline-profile updates memory-bank.md ## Offline profile and writes offline-profile.json to the project root. Step 13 (final summary) reads these for the wrap-up summary.

Hooked into /add-dataverse: Step 6 (just before this step) sets IsAvailableOffline=true + ChangeTrackingEnabled=true on every new custom table (see skills/add-dataverse/SKILL.md Step 5b body skeleton). That means /setup-offline-profile Gate 1 will be a no-op pass-through for fresh scaffolds — saving the user one approval click.

Step 7 — Auth config

Print before starting:

"→ [Step 7/13] Configuring app authentication (Entra ID app registration)…"

The template ships auth.config.json with blank msal.clientId and msal.tenantId. There are no baked-in registration IDs to reuse. Always use the selected Power Platform environment tenant resolved earlier in the flow, then ask the user how they want to provide the Entra app registration client ID.

auth.config.json may also contain a non-secret sibling environment object written by scripts/resolve-environment.js:

{
  "msal": { "clientId": "...", "tenantId": "..." },
  "environment": {
    "environmentId": "<guid>",
    "environmentUrl": "https://orgXXX.crm.dynamics.com",
    "tenantId": "<guid>",
    "cachedAt": "<iso timestamp>"
  }
}

Keep this block when editing auth.config.json. It lets later skills avoid re-running the environment-specific Power Platform API. Do not store tokens, secrets, or current-user Dataverse identity fields there.

7.1 Resolve selected Power Platform tenant

Use the environment selected in Step 4. Prefer $ACTIVE_TENANT_ID, then .resolved-environment.json, then the cached auth.config.json.environment.tenantId. Do not use the old msal.tenantId as an authority source; it may be blank or stale from a previous registration.

TENANT_ID="$ACTIVE_TENANT_ID"
if [ -z "$TENANT_ID" ]; then
  TENANT_ID=$(node -e "const j=require('./.resolved-environment.json'); console.log(j.tenantId || '')" 2>/dev/null || true)
fi
if [ -z "$TENANT_ID" ]; then
  TENANT_ID=$(node -e "const j=require('./auth.config.json'); console.log((j.environment && j.environment.tenantId) || '')" 2>/dev/null || true)
fi
echo "$TENANT_ID"

If TENANT_ID is empty, rerun scripts/resolve-environment.js using the environment ID in power.config.json:

ENV_ID=$(node -e "console.log(require('./power.config.json').environmentId || '')")
node "${PLUGIN_ROOT}/scripts/resolve-environment.js" "$ENV_ID" > .resolved-environment.json
TENANT_ID=$(node -e "const j=require('./.resolved-environment.json'); console.log(j.tenantId || '')")

If TENANT_ID is still empty, STOP and ask the user to fix environment resolution before continuing. Do not guess the tenant and do not copy msal.tenantId from auth.config.json.

7.2 Choose app registration path

Ask one question, using the resolved tenant:

"This app needs an Entra ID app registration in tenant <tenant-guid> to sign in.

Choose one: (a) Paste an existing app registration client ID (b) Create a new app registration from the Power Apps Wrap page, then paste its client ID (c) Skip for now — configure auth later"

Do not default to any option silently. The user must choose because app registration ownership varies by tenant/admin role.

• (a) Paste existing — run the client-ID write path in 7.3.
• (b) Create new in Power Apps Wrap — print the environment-specific Wrap URL in 7.4, then ask for the client ID and run 7.3. If the user cannot finish creation, allow skip and follow 7.5.
• (c) Skip — run the skip path in 7.5.

7.3 Write client ID into auth.config.json

Ask:

"Paste the Entra ID app registration client ID for tenant <tenant-guid> (GUID format), or type skip to configure auth later:"

If the user types skip, run 7.5. Otherwise validate UUID format. Write auth.config.json using Edit:

• Replace msal.clientId with the user's value
• Replace msal.tenantId with <tenant-guid> from 7.1
• Preserve the existing top-level environment block if present. If it is missing but .resolved-environment.json exists, add that JSON as top-level environment.

Do not create or modify the registration from this skill. The user owns it. Just wire the IDs into auth.config.json.

Print:

"→ Wired app registration into auth.config.json. Client ID: <id> Tenant: <tenant-guid>"

Jump to Step 8.

7.4 Create a new app registration in Power Apps Wrap

Resolve the selected Power Platform environment ID from $ACTIVE_ENV_ID, then power.config.json. Print the public Power Apps Wrap URL:

"Open the Power Apps Wrap app-registration page for the selected environment: https://make.powerapps.com/environments/<environment-id>/wraps#create-app-registration

Create the app registration on that page, then copy the Application (client) ID and paste it here. The Wrap experience configures the native registration for this flow. Do not add redirect URIs or API permissions manually; tenant-wide admin consent is not required. If you cannot create it now, type skip and run /set-app-registration-native later."

Tell the user the registration must be created/configured from the Power Apps Wrap page for the selected environment. Do not direct them to the Entra admin center for manual redirect URI, delegated permission, or admin-consent setup.

After the user creates the registration, run 7.3 to capture and write the client ID.

7.5 Skip auth for later

Write auth.config.json using Edit:

• Set msal.tenantId to <tenant-guid> from 7.1
• Leave msal.clientId as ""
• Preserve or add the top-level environment block from .resolved-environment.json if available

Print:

⚠️ Auth client ID is not configured. The app will fail to sign in until you add one. Run /set-app-registration-native later, or paste an app registration client ID into auth.config.json for tenant <tenant-guid>.

Do NOT touch src/playerConfig.ts — auth identifiers live in auth.config.json only.

Step 8 — Apply data model

Telemetry checkpoint: data_model

If <dataverse_planning_mode> = connector-only, verify the approved ## Data Model says zero Dataverse tables and no .datamodel-manifest.json exists, print ↷ Step 8 skipped — connector-only app has no Dataverse data model., skip Step 8.5 as well, and continue to Step 9. A non-empty Dataverse plan in this mode is a planning mismatch and must be corrected before continuing.

Print before starting:

"→ [Step 8/13] Preparing the approved Dataverse operation manifest, then invoking /add-dataverse for sequential metadata writes and service generation. Dataverse write time varies by environment; local manifest preparation is deterministic, not a wall-clock promise."

Environment pre-check (before invoking /add-dataverse): Verify that .resolved-environment.json / power.config.json match the environment captured in Step 1. If they differ, warn the user immediately — creating tables in the wrong environment is the #1 silent breakage in this step. /add-dataverse Step 3a does its own check, but catching it here saves a failed attempt.

For the fast-v2 required path, keep the foreground planning snapshot as planning evidence only. It never authorizes a write. Without changing any approval gate, Step 8 must perform one fresh bounded reconciliation for every exact table in the approved structured schema; each existing table reloads ordinary typed columns, lookups, M:N/1:N relationships, and alternate keys. Create/adapt table reruns also reload and compare creation-significant table behavior: ownership, activities, notes, offline availability, change tracking, labels/schema identity, and primary-name identity. The exact/proposed scope also includes every effective M:N intersect entity name, and 1:N reuse requires complete matching CascadeConfiguration evidence. An absent or colliding intersect name, or missing/mismatched cascade evidence, is non-executable. Step 8 also binds the structured artifact to the current fully approved plan content hash and the gate-owned approval receipt's exact contract hash and service dependencies. Step 8 cannot create or refresh this receipt. Use resolved context and these structured artifacts, never values inferred from free-form Markdown:

SCHEMA_CONTRACT="<working_dir>/.tmp/dataverse-schema-contract.json"
APPROVAL_RECEIPT="<working_dir>/.tmp/mobile-plan-status.json"
FOREGROUND_PLANNING_SNAPSHOT="<working_dir>/.tmp/dataverse-foreground-planning-snapshot.json"
RECONCILIATION_SCOPE="<working_dir>/.tmp/dataverse-reconciliation-scope.json"
EXECUTION_RECONCILIATION="<working_dir>/.tmp/dataverse-execution-reconciliation.json"
OPERATION_MANIFEST="<working_dir>/.tmp/dataverse-operation-manifest.json"
PUBLISH_CHECKPOINT="<working_dir>/.tmp/dataverse-publish-pending.json"
ACTIVE_SOLUTION_UNIQUE_NAME="Default"

test -f "$SCHEMA_CONTRACT" -a -f "$APPROVAL_RECEIPT" \
  -a -f "$FOREGROUND_PLANNING_SNAPSHOT" \
  -a -f "<working_dir>/native-app-plan.md"

node "${PLUGIN_ROOT}/scripts/build-dataverse-operation-manifest.js" \
  --bind-plan "$SCHEMA_CONTRACT" \
  --approval-receipt "$APPROVAL_RECEIPT" \
  --plan "<working_dir>/native-app-plan.md" \
  --output "$SCHEMA_CONTRACT"

node "${PLUGIN_ROOT}/scripts/build-dataverse-operation-manifest.js" \
  --reconciliation-scope "$SCHEMA_CONTRACT" \
  --output "$RECONCILIATION_SCOPE"

EXACT_TABLES=$(node -e "console.log(require(process.argv[1]).exactTables.join(','))" "$RECONCILIATION_SCOPE")
PROPOSED_TABLES=$(node -e "console.log(require(process.argv[1]).proposedTables.join(','))" "$RECONCILIATION_SCOPE")

node "${PLUGIN_ROOT}/scripts/create-dataverse-snapshot.js" \
  --env-url "$ACTIVE_ENV_URL" \
  --tenant-id "$ACTIVE_TENANT_ID" \
  --solution "$ACTIVE_SOLUTION_UNIQUE_NAME" \
  --tables "$EXACT_TABLES" \
  --proposed-tables "$PROPOSED_TABLES" \
  --reconcile-exact \
  --output "$EXECUTION_RECONCILIATION"

node "${PLUGIN_ROOT}/scripts/build-dataverse-operation-manifest.js" \
  --contract "$SCHEMA_CONTRACT" \
  --approval-receipt "$APPROVAL_RECEIPT" \
  --reconciliation "$EXECUTION_RECONCILIATION" \
  --plan "<working_dir>/native-app-plan.md" \
  --output "$OPERATION_MANIFEST" \
  --environment-id "$ACTIVE_ENV_ID" \
  --env-url "$ACTIVE_ENV_URL" \
  --tenant-id "$ACTIVE_TENANT_ID" \
  --publisher-prefix "$DETECTED_PUBLISHER_PREFIX" \
  --solution "$ACTIVE_SOLUTION_UNIQUE_NAME" \
  --publish-checkpoint "$PUBLISH_CHECKPOINT"

The manifest builder mechanically verifies the approved Reuse, Extend, Create, Adapt, Defer, and Unverified decisions. It must not invent or change architecture decisions. Any mismatch is a non-executable verification conflict and returns to the orchestrator; do not add another opportunistic read loop or fall back to agent reconciliation. No operation may execute until:

node "${PLUGIN_ROOT}/scripts/build-dataverse-operation-manifest.js" \
  --validate "$OPERATION_MANIFEST" \
  --contract "$SCHEMA_CONTRACT" \
  --approval-receipt "$APPROVAL_RECEIPT" \
  --reconciliation "$EXECUTION_RECONCILIATION" \
  --plan "<working_dir>/native-app-plan.md" \
  --environment-id "$ACTIVE_ENV_ID" \
  --env-url "$ACTIVE_ENV_URL" \
  --tenant-id "$ACTIVE_TENANT_ID" \
  --publisher-prefix "$DETECTED_PUBLISHER_PREFIX" \
  --solution "$ACTIVE_SOLUTION_UNIQUE_NAME" \
  --publish-checkpoint "$PUBLISH_CHECKPOINT" \
  --require-executable

Invoke /add-dataverse with the working directory, approved plan, and exact artifact paths:

Invoke skill: /add-dataverse

Arguments:
  --working-dir <working_dir>
  --plan-section <native-app-plan.md#data-model>
  --schema-contract <working_dir>/.tmp/dataverse-schema-contract.json
  --approval-receipt <working_dir>/.tmp/mobile-plan-status.json
  --execution-reconciliation <working_dir>/.tmp/dataverse-execution-reconciliation.json
  --operation-manifest <working_dir>/.tmp/dataverse-operation-manifest.json
  --publish-checkpoint <working_dir>/.tmp/dataverse-publish-pending.json
  --skip-planning   (the planner already ran)

/add-dataverse validates the bindings and consumes valid phases immediately. Any supplied artifact/binding failure returns to this orchestrator; it must not silently enter standalone reconciliation. A non-executable manifest authorizes no metadata writes. The publish checkpoint is retained across schema/PublishXml failure and deleted only after successful publish, so a rerun retries pending publication even when schema writes are already idempotent. It creates Tier 0 → N tables, applies extensions, runs npx power-apps add-data-source --api-id dataverse --org-url <envUrl> --resource-name <name> per service-required table from the app root, type-checks, and returns. Real matched A/B runs are still required to quantify the end-to-end time saved; do not present local manifest timing as a guaranteed 1–3 minute Dataverse result.

After /add-dataverse returns, run the Dataverse/generated-services gate:

npm run generate-schemas
npx tsc --noEmit

If this fails, do not continue to native capabilities, connectors, navigation, or screens. Capture the full error list once, batch-fix generated-service/model or alias-map issues, then rerun the gate. If the failure is a hidden Dataverse collision already recovered via an alias (for example aircraft → aircraftv2), make sure the alias is reflected in native-app-plan.md, memory-bank.md, and the Generated Services snapshot before rerunning.

Step 8.5 — Seed sample data (auto)

Print before starting:

"→ [Step 8.5/13] Checking existing record counts and seeding sample data into tables with fewer than 5 records."

Invoke /add-sample-data after Step 8. This step is not optional — every fresh-scaffolded app must have data to render on first launch:

Invoke skill: /add-sample-data

Arguments:
  --working-dir <working_dir>

/add-sample-data reads .datamodel-manifest.json, queries the current record count for each table, skips any table that already has ≥5 records, and seeds the rest with contextually appropriate rows in dependency-tier order. Inserted GUIDs are tracked in memory-bank.md for idempotent re-runs.

If .datamodel-manifest.json is missing, surface as DONE_WITH_CONCERNS and continue — it means Step 8 didn't complete cleanly.

If the seeding step fails (network drop, permission error, etc.), surface the failure but continue to Step 9 — the app is still usable, just empty on first launch. The user can re-run /add-sample-data later to retry.

Step 9 — Apply native capabilities

Print before starting:

"→ [Step 9/13] Wiring native capabilities: . Each runs sequentially."

Read the ## Native Capabilities section from native-app-plan.md. For each capability, invoke /add-native — it routes to nested helpers for camera/PDF/pen controls when needed, otherwise generates a generic wrapper:

Invoke skill: /add-native

Arguments:
  --working-dir <working_dir>
  --capability <name>

Run sequentially. Each writes a single file under src/native/ and does not touch package.json or app.config.js, so they could in principle run in parallel — but sequential keeps the orchestration log readable.

If the plan says "None — this app uses only standard React Native components and Power Platform connectors", skip only the native-capability invocation above and continue to Step 9a. Do NOT skip Step 9a or Step 9b; an app can need a pure-JavaScript library without any native capability, and Tamagui aliases/brand tokens are always required.

Step 9a — Install approved pure-JavaScript dependencies

Read and execute the Installation Contract in shared/references/javascript-dependency-planning.md for every approved row in ## Screens → ### JavaScript Dependencies. If the subsection is absent or says None., continue without changing dependencies.

Gate 4b approval is consent for exactly the packages and versions in the table. Install them into <working_dir> before any skeleton or builder imports them, validate package.json and the lockfile, and verify module resolution. Do not substitute another package/version, infer a package from a compiler error, or route a JS-only package through /add-native. If final inspection finds native code/config or incompatible runtime dependencies, remove only the newly added package and STOP with the exact failed criterion.

Step 9b — Apply design system

/design-system owns user-facing brand/design choices. The native-host createPowerAppsTamaguiConfig factory owns the baseline semantic aliases, contrast handling, animations, and font fallback. This step applies generated brand tokens without copying that host logic into the app.

Read the ## Design section from native-app-plan.md and follow the execution mapping in shared/references/design-planning.md:

Condition	Action
brand/tokens.ts exists	Highest priority. Apply ../design-system/references/tamagui-integration.md in brand-import mode, export the resolved app light/dark themes, then wire matching ThemeTokens into app/_layout.tsx.
## Design says required	Apply the same reference using the approved ## Design section. Builds custom token system + aliases.
## Design says add-aliases	Verify tamagui.config.ts uses createPowerAppsTamaguiConfig; the host already provides the semantic aliases.
Custom font only	npx expo install expo-font + useFonts() in _layout.tsx; preserve the host config factory.

No unchecked path. Screen-builders require $surface0–$surface3 and $accent* aliases. On the default path, verify the host factory rather than rewriting the config. Pass the complete ## Design section verbatim — not a summary. Re-run npx tsc --noEmit after Tamagui config changes.

Brand-token wiring — when brand/tokens.ts exists, the Tamagui integration exports appLightTheme and appDarkTheme. Map those resolved semantic values into the host themes so useTheme() and useThemeTokens() cannot drift:

import {
  PowerAppsProvider,
  lightTheme as hostLightTheme,
  darkTheme as hostDarkTheme,
} from '@microsoft/power-apps-native-host';
import type { ThemeTokens } from '@microsoft/power-apps-native-host';
import tamaguiConfig, {
  appDarkTheme,
  appLightTheme,
} from '../tamagui.config';

const brandedLightTheme: ThemeTokens = {
  ...hostLightTheme,
  surface0: appLightTheme.surface0,
  surface1: appLightTheme.surface1,
  surface2: appLightTheme.surface2,
  surface3: appLightTheme.surface3,
  surface4: appLightTheme.color6,
  text0: appLightTheme.text0,
  text1: appLightTheme.text1,
  text2: appLightTheme.text2,
  text3: appLightTheme.text3,
  accentDeep: appLightTheme.accentDeep,
  accentBase: appLightTheme.accentBase,
  accentSoft: appLightTheme.accentSoft,
  accentOnAccent: appLightTheme.accentOnAccent,
};
const brandedDarkTheme: ThemeTokens = {
  ...hostDarkTheme,
  surface0: appDarkTheme.surface0,
  surface1: appDarkTheme.surface1,
  surface2: appDarkTheme.surface2,
  surface3: appDarkTheme.surface3,
  surface4: appDarkTheme.color6,
  text0: appDarkTheme.text0,
  text1: appDarkTheme.text1,
  text2: appDarkTheme.text2,
  text3: appDarkTheme.text3,
  accentDeep: appDarkTheme.accentDeep,
  accentBase: appDarkTheme.accentBase,
  accentSoft: appDarkTheme.accentSoft,
  accentOnAccent: appDarkTheme.accentOnAccent,
};

// In RootLayout:
<PowerAppsProvider ... theme={brandedLightTheme} darkTheme={brandedDarkTheme}>

The generated schema has one brand palette, so the exported dark app theme retains Config v5 dark surfaces and text while carrying brand accents and statuses. For runtime theme switching, use useThemeControl() from @microsoft/power-apps-native-host.

Step 10 — Add connectors

Print before starting:

"→ [Step 10/13] Adding connectors: . Each runs sequentially (parallel writes would race)."

Read the ## Connectors section from native-app-plan.md. If it says "None", skip this step entirely.

For each row in the table, route to the correct skill based on the API name:

API name	Invoke
sharepointonline	/add-sharepoint --working-dir <working_dir>
anything else	/add-connector --working-dir <working_dir> --connector <api-name>

Run sequentially — each generates files under src/generated/. Parallel writes would race.

Mutation-heavy steps stay sequential. Dataverse table creation (Step 8), connector adds (Step 10), and generated-service writes are all sequential by design. The fast path in this skill is parallel screen generation (Step 11) plus fewer prompts (token cache, sticky policies, auto-proceed) — NOT parallelizing the data-source/service mutations. Do not attempt to parallel-batch npx power-apps add-data-source or /add-connector invocations; they share src/generated/ and power.config.json and will race or corrupt state.

Step 10b — Wire navigation layout

Read ## Screens → Navigation Pattern from native-app-plan.md.

• Stack — skip. app/(app)/_layout.tsx already renders <Stack>. Nothing to do.
• Tabs or Tabs + Stack — write outer <Tabs> in app/(app)/_layout.tsx AND a per-folder inner <Stack> in each app/(app)/<folder>/_layout.tsx.
• Drawer — write outer <Drawer> in app/(app)/_layout.tsx AND a per-folder inner <Stack> in each app/(app)/<folder>/_layout.tsx.

⚠️ The phantom-tab fix lives here. expo-router auto-registers every top-level .tsx file under app/(app)/ as a tab/drawer entry. Step 10b prevents phantom entries by walking the File column in the Screen Map (not the Screen names): each unique top-level entry under app/(app)/ — file OR folder — becomes ONE tab/drawer entry. Folders contain detail/modal screens inside their own stack, so they never leak as siblings.

Step 10b.1 — Compute the layout structure from the Screen Map

Read the Screen Map's File column. For every row whose File starts with app/(app)/, classify each path into one of three groups:

Path shape	Classification	Example
app/(app)/<name>.tsx (no subfolder)	Top-level flat file — one outer entry, no inner layout	app/(app)/home.tsx
app/(app)/<folder>/index.tsx	Folder root — one outer entry, needs inner _layout.tsx	app/(app)/inspections/index.tsx
app/(app)/<folder>/<child>.tsx (any other file inside a folder)	Folder child — pushed into the folder's stack, NOT an outer entry	app/(app)/inspections/[id].tsx

Build two lists from the classification:

1. Outer entries = unique <name> from the top-level flat files + unique <folder> from the folder roots. These get one <Tabs.Screen> or <Drawer.Screen> each in the outer layout.
2. Inner stacks = one entry per unique <folder>. For each folder, list its children (root + non-root files), with each child's Presentation value from the Screen Map.

Sanity check before writing anything: if any folder has children but no index.tsx row in the Screen Map, STOP and report: BLOCKED: folder app/(app)/<folder>/ has children (<list>) but no index.tsx row in the Screen Map. The screen-planner must emit an index.tsx row for every folder. This catches a planner mistake that would render the folder unreachable from the outer tab.

Normalize every Screen Map file to its Expo route (strip .tsx, collapse trailing /index, preserve dynamic segments). If two files normalize to the same route, STOP before writing layouts. In particular, reject <parent>/[id].tsx together with <parent>/[id]/<child>.tsx; move the detail contract to <parent>/[id]/index.tsx.

BLOCKED: duplicate Expo route <route> from <file-a> and <file-b>. Use [id]/index.tsx when a dynamic detail route owns child screens.

Step 10b.2 — Write per-folder inner _layout.tsx files (if any folders exist)

For each entry in the Inner stacks list, create the folder if missing and write app/(app)/<folder>/_layout.tsx with this template:

import { Stack } from 'expo-router';

export default function <FolderName>Layout() {
  return (
    <Stack screenOptions={{ headerShown: false }}>
      <Stack.Screen name="index" />
      {/* one <Stack.Screen> per non-index child, with presentation from Screen Map */}
      <Stack.Screen name="<child-without-tsx>" options={{ presentation: '<Presentation>' }} />
    </Stack>
  );
}

Rules:

• headerShown: false at the Stack level — each screen sets its own header inline via <Stack.Screen options={{...}}> at the top of its component (the Expo Router idiom).
• <Stack.Screen name="index" /> is required — without it, the folder root won't render.
• presentation: 'modal' and presentation: 'formSheet' come from the Screen Map's Presentation column. Skip the options prop entirely for default presentation.
• name for [id].tsx is literally [id] (with brackets). When [id] owns child routes, create <folder>/[id]/_layout.tsx with <Stack.Screen name="index" /> and child entries; do not register both [id].tsx and a [id]/ folder.
• Folder name in the function name is PascalCase (e.g. InspectionsLayout).

Why this must run BEFORE Step 11: screen-builders write their files in parallel, multiple builders may target the same folder, and any of them creating _layout.tsx would race. The orchestrator owns these files.

Step 10b.3 — Write outer app/(app)/_layout.tsx

Now rewrite only the return statement in app/(app)/_layout.tsx. Keep every line above the return untouched (auth guard, all imports).

How to build the <Tabs> block (Tabs / Tabs + Stack pattern):

For each entry in the Outer entries list, emit one <Tabs.Screen>. The name is the file/folder name without .tsx:

For each tab, infer a Ionicons icon name from the screen name:

Screen name contains	Icon
home, dashboard, overview	home-outline
inspect, audit, checklist, task	clipboard-outline
profile, account, me, user	person-outline
settings, config, preferences	settings-outline
report, analytics, chart, stats	bar-chart-outline
map, location, sites, field	map-outline
message, chat, inbox, notify	chatbubble-outline
anything else	apps-outline

The Edit to apply:

Add import { Tabs } from 'expo-router';, import { Ionicons } from '@expo/vector-icons';, and import { useThemeTokens } from '@microsoft/power-apps-native-host'; to the import block if not already present. Inside AppLayout, after the auth state is read, add const theme = useThemeTokens();. Then replace:

return (
  <Stack
    screenOptions={{
      headerShown: false,
    }}
  />
);

with:

return (
  <Tabs
    screenOptions={{
      headerShown: false,
      tabBarActiveTintColor: theme.accentBase,
      tabBarInactiveTintColor: theme.text2,
    }}
  >
    <Tabs.Screen
      name="<screen-file-name>"
      options={{
        title: '<Screen Title>',
        tabBarIcon: ({ color }) => <Ionicons name="<icon>" size={22} color={color} />,
      }}
    />
    {/* one Tabs.Screen per top-level tab */}
  </Tabs>
);

Run npx tsc --noEmit after the edit. If it fails, check that the Tabs.Screen name values exactly match the file names under app/(app)/ (without .tsx).

How to build the <Drawer> block (Drawer pattern only):

Same Outer-entries computation as Tabs — one entry per top-level flat file or folder root from Step 10b.1. Detail, modal, and nested routes are inside their folder's inner stack, not drawer items.

Use the same icon mapping table as Tabs (above).

The Edit to apply:

Add import { Drawer } from 'expo-router/drawer';, import { Ionicons } from '@expo/vector-icons';, and import { useThemeTokens } from '@microsoft/power-apps-native-host'; to the import block if not already present. Inside AppLayout, after the auth state is read, add const theme = useThemeTokens();. Then replace the existing <Stack> return with:

return (
  <Drawer
    screenOptions={{
      headerShown: true,
      drawerType: 'front',
      drawerActiveTintColor: theme.accentBase,
      drawerInactiveTintColor: theme.text2,
      drawerStyle: { width: 280 },
    }}
  >
    <Drawer.Screen
      name="<screen-file-name>"
      options={{
        title: '<Screen Title>',
        drawerIcon: ({ color }) => <Ionicons name="<icon>" size={22} color={color} />,
      }}
    />
    {/* one Drawer.Screen per top-level destination */}
  </Drawer>
);

Key differences from Tabs:

• Import is from 'expo-router/drawer' (not from 'expo-router')
• headerShown: true — drawer needs the hamburger icon in the header; hiding it makes the drawer unreachable
• drawerType: 'front' — standard mobile pattern (drawer slides over content)
• Icon prop is drawerIcon (not tabBarIcon)

Run npx tsc --noEmit after the edit. If it fails, check that the Drawer.Screen name values exactly match the file names under app/(app)/ (without .tsx).

Step 10.7 — Snapshot generated services into the plan

Print before starting:

"→ [Step 10.7/13] Probing src/generated/services/ and writing the service registry into native-app-plan.md…"

Before spawning N parallel screen-builders, the orchestrator probes src/generated/services/ ONCE and writes the result into native-app-plan.md. Without this, every builder runs its own Glob, may spell service names differently, and ends up with mixed states inside one app (some screens use CrXxxService.getAll(), others write // TODO(connector-not-yet-added) for the same service).

cd <working_dir>
ls -1 src/generated/services/*.ts 2>/dev/null | sed 's|src/generated/services/||;s|\.ts$||'

For each service file found, run a quick grep to list its exported methods so builders know what's actually available without re-reading the (large) generated file:

for svc in $(ls -1 src/generated/services/*.ts 2>/dev/null); do
  name=$(basename "$svc" .ts)
  methods=$(grep -oE 'static async [a-zA-Z_]+' "$svc" | sed 's/static async //' | tr '\n' ',' | sed 's/,$//')
  echo "| \`$name\` | \`src/generated/services/$name.ts\` | $methods |"
done

Write the result into native-app-plan.md as a new section immediately after ## Screens (and refresh it on every re-run — services come and go as the user runs /add-dataverse, /add-sharepoint, etc.):

## Generated Services (snapshot at <ISO timestamp>)

| Service | Path | Methods present |
|---|---|---|
| `Cr3e9_projectsService` | `src/generated/services/Cr3e9_projectsService.ts` | `getAll, get, create, update, delete` |
| `Cr3e9_tasksService` | `src/generated/services/Cr3e9_tasksService.ts` | `getAll, get, create, update, delete` |

**For screen-builders:** if a service your spec references is in this table, import it and use the exact name + methods listed. If it is NOT in this table, the data source has not been added yet — write the screen with the expected import path and a `// TODO(connector-not-yet-added): run /add-dataverse to generate <ServiceName>` comment so the user can see what's blocked. Do not invent or rename services.

If the directory is empty (no data sources added yet), still write the section with an empty table and a one-line note: "No generated services yet — builders will emit TODO stubs for any service their spec references."

Step 10.8 — Generate app-specific shared code + screen skeletons

Print before starting:

"→ [Step 10.8/13] Generating app-specific components, hooks, utils, and screen skeletons from the plan…"

This step analyzes the per-screen specs and generates shared code that multiple screens will use plus typed skeleton files for each screen. Builders then fill in the JSX rather than starting from zero. This cuts builder output by ~50% and eliminates import-path guessing errors.

---

10.8a — Analyze plan for cross-screen patterns

Read all per-screen specs in ## Screens → ### Per-Screen Specs. Identify:

1. Entity cards/rows — if 2+ screens render the same entity (same Service) in a card/row format, generate a shared component.
2. Choice column maps — if 2+ screens reference the same choice column (e.g. status: 1=Pending, 2=Active), generate a constants file.
3. Custom hooks — if 2+ screens call the same service with similar params (e.g. both list + detail call InspectionsService), generate a domain hook.
4. Shared formatters — if screens need entity-specific formatting (e.g. "inspection title" = ${name} · ${equipment}), generate a formatter.

Decision rules:

Pattern in specs	Generate	Where
Same entity shown as list-item on 2+ screens	<Entity>Card.tsx or <Entity>Row.tsx	src/components/
Same choice column referenced on 2+ screens	constants.ts with ENTITY_STATUS map + tone mapping	src/utils/
Same bounded service + similar .getAll() params on 2+ screens	use<Entity>List.ts wrapping useListData	src/hooks/
Same cursor-paginated service on 1+ unbounded screens	use<Entity>CursorList.ts wrapping useCursorListData	src/hooks/
Entity detail + edit screens for same entity	use<Entity>.ts with get + save + delete	src/hooks/

Write the files directly into the project (not into samples — these are app-specific):

# Example — if plan has "Inspections" entity used on list + detail + home screens:
cat > "<working_dir>/src/components/InspectionRow.tsx" << 'EOF'
... generated component ...
EOF

If no cross-screen patterns are found (e.g. only 2 screens total with no overlap), skip this sub-step — the shared scaffold is sufficient.

---

10.8b — Generate screen skeletons

For each screen in the plan's Screen Map that will be built by a screen-builder, write a typed skeleton file at its target_file path. The skeleton contains:

1. All imports (components, hooks, utils, services, types) pre-resolved
2. The exported component function with typed props/params
3. The hook calls (e.g. useListData, useSearchFilter, useLocalSearchParams)
4. An empty return with a // TODO: screen-builder fills JSX here marker

Skeleton template for a Cursor List screen (Pagination: cursor):

import React from 'react';
import { FlatList, RefreshControl } from 'react-native';
import { useRouter } from 'expo-router';
import { YStack, XStack, Text, Input, Spinner } from 'tamagui';
import { Ionicons } from '@expo/vector-icons';
import { SafeAreaView } from 'react-native-safe-area-context';
import { LoadingState, ErrorState, EmptyState, ScreenHeader } from '@/components';
import { useCursorListData } from '@/hooks';
import { containsFilter, formatDate, choiceLabel } from '@/utils';
import { <Service> } from '@/generated/services/<Service>';
import type { <Entity> } from '@/generated/models/<Entity>Model';
// App-specific imports (if generated at 10.8a):
import { <Entity>Row } from '@/components/<Entity>Row';
import { <ENTITY>_STATUS } from '@/utils/constants';

export default function <ScreenName>() {
  const router = useRouter();
  const { items, loading, refreshing, loadingMore, hasNextPage, error, query, setQuery, onRefresh, refetch, loadMore } = useCursorListData<<Entity>>({
    queryKey: ['<entityPlural>'],
    fetchPage: ({ pageSize, search, skipToken }) => <Service>.getAll({
      maxPageSize: pageSize,
      orderBy: ['<orderField> desc', '<primaryKey> asc'],
      select: [<renderedColumns>],
      ...(search ? { filter: containsFilter('<searchColumn>', search) } : {}),
      ...(skipToken ? { skipToken } : {}),
    } as any),
  });

  // TODO: screen-builder fills JSX here. FlatList MUST wire:
  // - data={items}
  // - refreshControl={<RefreshControl refreshing={refreshing} onRefresh={onRefresh} />}
  // - onEndReached={hasNextPage ? loadMore : undefined}
  // - ListFooterComponent={loadingMore ? <Spinner /> : null}
  return null;
}

Skeleton template for a Bounded List screen (Pagination: none):

import React from 'react';
import { FlatList } from 'react-native';
import { useRouter } from 'expo-router';
import { YStack, XStack, Text, Input } from 'tamagui';
import { Ionicons } from '@expo/vector-icons';
import { SafeAreaView } from 'react-native-safe-area-context';
import { LoadingState, ErrorState, EmptyState, ScreenHeader } from '@/components';
import { RefreshControl } from 'react-native';
import { useListData, useSearchFilter } from '@/hooks';
import { formatDate, choiceLabel } from '@/utils';
import { <Service> } from '@/generated/services/<Service>';
import type { <Entity> } from '@/generated/models/<Entity>Model';
// App-specific imports (if generated at 10.8a):
import { <Entity>Row } from '@/components/<Entity>Row';
import { <ENTITY>_STATUS } from '@/utils/constants';

export default function <ScreenName>() {
  const router = useRouter();
  const { items, loading, refreshing, error, onRefresh, refetch } = useListData(
    () => <Service>.getAll({ orderBy: ['<orderField> desc'], top: 50 }),
  );
  const { query, setQuery, filtered } = useSearchFilter(items, [<searchKeys>]);

  // TODO: screen-builder fills JSX here
  return null;
}

Do NOT use the bounded skeleton for a screen whose spec says Pagination: cursor. useListData fetches one bounded page; useSearchFilter filters only loaded rows. Cursor screens must use useCursorListData, useInfiniteQuery, or an app-specific cursor hook generated in 10.8a.

Skeleton template for a Detail screen:

import React from 'react';
import { ScrollView } from 'react-native';
import { useLocalSearchParams, useRouter } from 'expo-router';
import { YStack, XStack, Text, Button } from 'tamagui';
import { Ionicons } from '@expo/vector-icons';
import { SafeAreaView } from 'react-native-safe-area-context';
import { LoadingState, ErrorState, BottomActionBar, InfoRow } from '@/components';
import { formatDate, choiceLabel } from '@/utils';
import { <Service> } from '@/generated/services/<Service>';
import type { <Entity> } from '@/generated/models/<Entity>Model';

export default function <ScreenName>() {
  const { id } = useLocalSearchParams<{ id: string }>();
  const router = useRouter();
  const [item, setItem] = React.useState<<Entity> | null>(null);
  const [loading, setLoading] = React.useState(true);

  React.useEffect(() => {
    if (!id) return;
    <Service>.get(id).then(r => { setItem(r.data ?? null); setLoading(false); });
  }, [id]);

  // TODO: screen-builder fills JSX here
  return null;
}

Skeleton template for a Form screen:

import React, { useState } from 'react';
import { Alert, ScrollView, KeyboardAvoidingView, Platform } from 'react-native';
import { useRouter, useLocalSearchParams } from 'expo-router';
import { YStack, Text, Button, Input } from 'tamagui';
import { ModalHeader, FormField, RowPick } from '@/components';
import { <Service> } from '@/generated/services/<Service>';

export default function <ScreenName>() {
  const router = useRouter();
  const [saving, setSaving] = useState(false);
  // Form state fields from spec:
  <field_declarations>

  const submit = async () => {
    // TODO: screen-builder fills validation + service call
  };

  // TODO: screen-builder fills JSX here
  return null;
}

Skeleton template for the Profile screen (/(app)/profile, always generated; app-specific content comes from the Profile spec):

import React from 'react';
import { Alert } from 'react-native';
import { useRouter } from 'expo-router';
import { YStack, XStack, Text, Button } from 'tamagui';
import { SafeAreaView } from 'react-native-safe-area-context';
import { useAuth } from '@microsoft/power-apps-native-host';

export default function <ScreenName>() {
  const router = useRouter();
  // AuthState shape (@microsoft/power-apps-native-host): { isLoading, isAuthReady, isSignedIn, error, acquireToken, signIn, signOut }
  // There is NO `user` / `account` field. Display name comes from the ID-token claim, not from useAuth().
  const { isSignedIn, signOut } = useAuth();

  const handleSignOut = React.useCallback(() => {
    Alert.alert('Sign out?', 'You can sign in again with your work or school account.', [
      { text: 'Cancel', style: 'cancel' },
      {
        text: 'Sign out',
        style: 'destructive',
        onPress: () => {
          void Promise.resolve()
            .then(() => signOut())
            .finally(() => router.replace('/login'));
        },
      },
    ]);
  }, [router, signOut]);

  // TODO: screen-builder fills JSX here. Any visible Sign out button calls handleSignOut.
  return null;
}

Rules for skeleton generation:

• Replace <Service>, <Entity>, <ScreenName>, <searchKeys>, <orderField>, <field_declarations> with actual values from the plan's per-screen spec + Generated Services table.
• If a service is NOT in the Generated Services table, still write the import but add // TODO(connector-not-yet-added) above it.
• Profile skeletons may also include generated service/model imports when the Profile spec's Profile content or Data fields require persisted app-specific user context. Keep the sign-out helper regardless of whether Profile also loads data.
• The skeleton is a valid TypeScript file (compiles with return null) — builders replace the return null with real JSX.
• Do NOT write skeletons for screens that already exist in the template (e.g. home.tsx if it's already present).
• The Profile skeleton must keep the handleSignOut helper and wire the planned Sign out button to it. Do not inline a second auth/logout path, and do not make Profile a sign-out-only screen.
• Do not merge sign-out imports or helpers into non-Profile screen skeletons. The Profile screen is the only sign-out owner.
• Never destructure user, account, profile, or claims from useAuth() — those fields do not exist on AuthState. The only fields are isLoading, isAuthReady, isSignedIn, error, acquireToken, signIn, signOut. If the screen needs the signed-in user's name/email, add a // TODO: decode ID token claim comment — do not invent a field.

---

10.8c — DEPRECATED (skeleton is the import source of truth)

This sub-step previously appended ### Standard Imports + per-screen #### Resolved Imports blocks into the plan. That added ~150 lines on a 14-screen plan and duplicated the imports already pre-resolved into each skeleton file at Step 10.8b.

The skeleton file IS now the single source of truth for per-screen imports + hook calls. The screen-builder reads the skeleton at target_file and fills the JSX. Do NOT append duplicate import documentation into the plan.

10.8d — Navigation/skeleton TypeScript gate

After Step 10b layouts, Step 10.7 service snapshot, and Step 10.8 shared code/skeletons are all written, run the Navigation/skeleton gate:

npx tsc --noEmit

If this fails, do not launch Step 11. Capture the full error list once, batch-fix layout names, route paths, skeleton imports, shared component exports, generated service imports, or hook signatures, then rerun the gate. Screen-builders should start only from a clean shell with typed skeletons that compile with return null.

Step 11 — Build screens (parallel)

Telemetry checkpoint: screens

Build mode is NEVER a user-facing question. Do not ask "Build mode? parallel/inline" or any variant. The orchestrator decides automatically per the preflight below.

Quality rule — screen count/time is NOT a fallback trigger. If Task can spawn mobile-app:screen-builder, always use screen-builder waves, even for 10+ screens. Do NOT write "given the scale/time, I'll write screens inline" or any equivalent shortcut. Screen-builder agents carry the quality checklist, domain-pattern rules, resolved-import discipline, safe-area/contrast/a11y checks, and per-screen return protocol. Inline mode exists only for host/tooling failure, not for convenience.

11.0 — Task preflight (silent)

Before the first wave, do a one-shot probe to confirm Task can spawn mobile-app:screen-builder. Spawn ONE no-op builder via Task with screen_name: __preflight__ and target_file: <working_dir>/.preflight-probe.tsx (the probe agent should return DONE immediately and the orchestrator deletes the file).

• Probe returns DONE (any status code) → parallel mode. Proceed with normal fan-out below. Do not downgrade later because there are many screens or because inline writing seems faster.
• Probe errors with Agent type … not found, tool unavailable, or any nested-Task failure → silent inline fallback. Print once:

  "→ Parallel screen-builders unavailable in this host — building screens inline. (No action needed; this is automatic.)" Then iterate the screen list and apply the full screen-builder workflow inline (the orchestrator becomes the builder, reading the same per-screen specs and writing TSX directly). Inline does NOT mean "concise but functional" — it must still satisfy screen-builder.md quality gates, operational pattern requirements, safe-area/contrast/a11y rules, resolved imports, and the final checklist for every screen. Do NOT prompt the user.

Hard rule — never ask the user about build mode. The probe is the only decision input. If the host changes mid-run (rare), treat the next failure the same way: silently downgrade to inline and continue.

Hard rule — no nested agent spawning. Screen-builder agents MUST NOT spawn further agents (no nested Task calls). The top-level orchestrator owns the entire screen-builder fan-out: one Task batch per wave of up to 5 screens. If a builder needs help that previously would have been a nested spawn, it returns NEEDS_CONTEXT: and the orchestrator handles the follow-up at the wave boundary.

Print before spawning (substitute computed values; <W> = total waves = ceil(N/5)):

"→ [Step 11/13] Building screens in wave(s) of up to 5 concurrent. Wave 1/ starting: ."

Read the ## Screens section's per-screen specs. For each screen the plan marks as new (skip baseline screens already in template), spawn a mobile-app:screen-builder agent via Task in a single message so they run in parallel. The mobile-app: plugin-name prefix is required.

Spawn N agents (parallel): mobile-app:screen-builder

Each prompt:
  working_dir: <working_dir>
  screen_name: <name>
  route: <route>
  target_file: <working_dir>/<File from Screen Map>
  plan_path: <working_dir>/native-app-plan.md
  skeleton_exists: true

  Follow screen-builder.md. Build from the user's compact per-screen spec, shared conventions, and design direction — inherited defaults are intentional, and samples are API/import references only, not layouts to copy. A typed skeleton already exists at your target_file with all imports and hook calls pre-resolved from the Generated Services table + per-screen `**Data**` field — fill in the JSX, do not discard imports. The skeleton file IS the import source of truth; the plan no longer documents per-screen imports separately. Return per AGENTS.md rule #10: literal first line is `DONE` / `DONE_WITH_CONCERNS:` / `NEEDS_CONTEXT:` / `BLOCKED:`, then a blank line, then the one-line summary.

target_file resolution (HARD): read the File column from the Screen Map row for this screen and prefix it with <working_dir>/. The path may be nested (e.g. <working_dir>/app/(app)/inspections/[id].tsx). The folder is guaranteed to exist because Step 10b.2 created it and wrote the inner _layout.tsx. Do NOT compute the path as <working_dir>/app/(app)/<screen-name>.tsx — that strips the folder structure and produces phantom-tab files. If the Screen Map row has no File column (older planner output), fall back to the flat path and surface a DONE_WITH_CONCERNS: Screen Map missing File column — used flat fallback paths, expect phantom tabs after the wave.

Cap at 5 concurrent. If the plan has more than 5 new screens, batch them in waves of 5.

Progress streaming — print one line per builder as the wave returns, then a wave summary. The Task tool returns all parallel results together, but you can still narrate per-builder by iterating the returned results in order before doing the status-switch branching. Format:

  ✓ [3/8] HomeScreen — DONE
  ✓ [4/8] ListScreen — DONE_WITH_CONCERNS (1 connector stub)
  ✓ [5/8] DetailScreen — DONE
─── Wave 1/2 complete (5/8 screens built; 0 blocked, 1 with concerns) ───

Use ✓ for DONE / DONE_WITH_CONCERNS, ↻ for NEEDS_CONTEXT (will retry), ✗ for BLOCKED. Always print the running counter [K/N] so the user sees forward motion. The wave summary line goes on its own line after the per-builder block.

After the wave's TypeScript gate passes, and only then, print the next wave start line (if any):

"Wave 2/ starting: ."

After each wave returns, run the Step 3.0 status switch on every builder's first line. Branch per builder:

• DONE → continue.
• DONE_WITH_CONCERNS: <list> (typical case: a // TODO(connector-not-yet-added) stub was emitted because the referenced service is not in the Generated Services table) → batch concerns across all builders, surface the consolidated list to the user once at the end of the wave (not per-builder — that would be noise), and ask whether to fix any pending connectors via /add-connector before continuing to Step 12. Record in memory-bank.md.
• NEEDS_CONTEXT: <missing> → re-spawn that one builder with the missing context appended to its prompt (cap 2 retries per screen, then BLOCKED). Print ↻ [K/N] <name> — retrying (missing: <missing>) so the user understands the wave isn't fully clean yet.
• BLOCKED: <reason> → STOP for that screen, print ✗ [K/N] <name> — BLOCKED (<reason>) and ask the user whether to (1) fix and retry, (2) skip the screen and continue with a placeholder, or (3) abort the whole flow.

After handling every builder status in the wave, run the Screen-wave gate before launching the next wave:

npx tsc --noEmit

If the wave gate fails, capture the full error list once, group failures by root cause, and repair in batch. For screen-owned files, re-spawn the affected screen-builder(s) with the consolidated TypeScript output appended to their prompts. Affected builders can be re-spawned in parallel. Cap retries at 2 per screen, then surface the failure to the user. Do not launch the next wave until the current wave gate is clean.

Common wave-gate repair classes to batch instead of fixing line-by-line:

• Generated service/model names: singular vs plural generated names, stale aliases after Dataverse rename.
• Service option shapes: orderBy must match the generated type, usually string[].
• UI prop mismatches: invalid Tamagui shorthand props on components that do not support them.
• React Native style types: percent widths must use a typed percentage or shared ProgressBar helper.
• Dataverse create/update payload typing: prefer typed helper wrappers; if generated base types require server-owned fields, isolate any as any at the helper boundary, not throughout screen JSX.
• Stale connector TODOs: remove TODO(connector-not-yet-added) when the service exists in the Generated Services snapshot.

After all waves return and the last wave gate is clean, run one final npx tsc --noEmit before Step 12 to catch cross-screen issues that only appear when all screens exist. If it fails, use the same consolidated batch-repair flow.

Then run the canonical route-contract gate from the app root:

node "${PLUGIN_ROOT}/scripts/check-routes.js"

This gate is required even when TypeScript passes. It detects duplicate normalized routes, [id].tsx plus [id]/<child>.tsx file/folder collisions, and sender/destination parameter drift. If it fails, repair the affected route files or re-spawn their screen builders with the consolidated findings, then rerun once. Do not continue to Step 11.4 or start Metro while route findings remain.

Sticky tsc/build error policy (run-level). The first time a tsc or npm run build failure surfaces in this run, ask the user once:

"tsc found error(s) in . Patch + continue, or stop and let me investigate?"

Record the answer in memory-bank.md under ## Policies as tsc_error_policy: patch_continue or tsc_error_policy: stop_for_review. For every subsequent tsc/build error of the same class in the same run (e.g., another screen failing typecheck after a builder retry, the cross-screen tsc after Step 11.4 fixes), apply the recorded policy automatically:

• patch_continue → re-spawn the matching builder with the error appended (or auto-patch in inline mode), respecting the 2-retry cap. Do not re-prompt the user.
• stop_for_review → STOP and surface the new error.

Reset the policy only if the user explicitly says "ask me again" or /edit-app is invoked. This avoids the same class of question being asked 3–5 times per run while still letting the user override at any point.

This sticky policy controls how to handle a failed gate, not whether the gate is required. Even with patch_continue, every required TypeScript gate must end clean before the flow advances.

Step 11.4 — Stylistic fix sweep (parallel)

Run one controlled stylistic debt sweep after all screen-builder waves and TypeScript gates are clean, before preview or dev-server launch. This keeps screen-builder retries focused on critical compile/data/route issues, then fixes visual and accessibility quality across the full screen set in batches.

Print before starting:

"→ [Step 11.4/13] Running stylistic validators in batch + auto-fixing contrast / accessibility / token issues across all screens (~2-3 min)"

Scope: generated screen files only: every file from the Screen Map plus any app/(app)/**/*.tsx screen written by Step 10.8/Step 11. Exclude layout files unless the reported issue is clearly inside generated screen chrome for that route group. Do not scan src/generated/, brand/, node_modules/, .expo/, or sample files.

Available validators in v0:

node "${PLUGIN_ROOT}/hooks/validate-screen-quality.js" --report <screen-files-or-app-dir>
node "${PLUGIN_ROOT}/hooks/validate-color-contrast.js" --report <screen-files-or-app-dir>

validate-screen-quality includes accessibility-label/role, safe-area, touch-target, raw-hex, token, empty-state, shadow, and status-visual checks. If future stylistic hooks exist (for example validate-accessibility-labels.js), include them here only if they support --report and emit the same JSON issue shape.

For each available stylistic validator:

1. Run in --report mode against all generated screens. Report mode is non-blocking; it emits JSON issues with file, line, rule, match, fix, and autoFixable.
2. Merge issues by file and rule. Keep exact line numbers for user/debug output, but do not rely on stale line numbers after the first edit in a file.
3. Split findings into deterministic auto-fixes and judgement calls:

• Auto-fixable: weak foreground tokens, white-on-yellow/orange status pairs, missing icon-only aria-label, missing tappable role, tiny icon button hitSlop, obvious raw hex/token substitutions, top-only safe area with bottom UI, allowFontScaling={false}. Apply these web-standard accessibility props to Tamagui 2 components; raw React Native components retain their React Native accessibility props.
• Needs review: complex safe-area restructuring, dominant red detail headers, redundant status cue design, ambiguous brand colors, empty-state restructuring that requires moving JSX across large blocks.

4. Build one file-level edit batch per affected file. Apply affected files in parallel because screen files are independent. Do not run one edit per issue when multiple issues are in the same file; that reintroduces slow per-write loops and line-number drift.
5. Re-run the same validator in --report mode for the touched files. Cap retries at 2 per file per validator.

These validators are invoked explicitly by this mobile workflow. They are not registered as plugin-wide hooks because that would run them during unrelated Canvas Apps and other plugin operations.

After all validators report no auto-fixable issues, run:

npx tsc --noEmit

If tsc fails, use the existing TypeScript batch-repair policy. If stylistic issues remain after 2 retries or are judgement calls, do not keep looping. Record them in memory-bank.md and surface them as:

DONE_WITH_CONCERNS: Step 11.4 left <N> stylistic issue(s) for review: <file:line rule summary>

Then continue only if TypeScript is clean. Step 11.4 may leave concerns, but it may not leave the app in a broken TypeScript state.

Optional static preview

After tsc passes, offer a static HTML preview. The dev server starts next (Step 12), so default is skip:

"→ N screens built and type-checked. The live app starts next.

Want a static HTML preview first, or go straight to the live app?

(a) Preview all screens — HTML phone frames for every screen (b) Preview key screens — List + Form + Detail archetypes only (c) Skip preview

[default: c]"

• (a) → invoke /preview-screens (all screens)
• (b) → invoke /preview-screens with only List + Form + Detail screen files (skip Login, Splash, Profile, OAuth)
• (c) → proceed directly to Step 12

---

Step 12 — Start dev server (background)

Telemetry checkpoint: app_ready

Print before starting:

"→ [Step 12/13] Launching Metro dev server in the background so you can scan the QR."

This skill launches Metro in an async/background terminal so:

1. The QR code prints in the terminal — the user can scan with their dev client immediately.
2. Hot-reload works on file edits — no restart needed for screen tweaks.
3. The agent owns the terminal — when the user says "the screen is blank" / "data isn't showing" / "it crashed", the agent can read Metro's console.log, BUNDLE errors, and red-box stack traces directly via BashOutput (or its equivalent terminal-output tool) without asking the user to copy-paste.

Launch commands:

cd <working_dir>
npm run generate-schemas    # refresh schema map for any data sources added since last run (idempotent)
npx tsc --noEmit            # final gate — dev server starts only from a clean TypeScript state

Run the schema regen and final tsc synchronously and check both exits. If either fails, do not launch Metro. Capture the full output once, batch-fix by root cause, rerun the final gate, and continue only when clean. Then launch Metro async:

# Async / background — DO NOT block on this. Capture the terminal id.
npx expo start

Use npx expo start here instead of npm run dev because the orchestrator has already run npm run generate-schemas for the final gate. The template keeps predev: npm run generate-schemas as a safety net for humans running npm run dev manually, but the orchestrated path should not regenerate schemas twice.

When invoking the Bash tool: set run_in_background: true (or the equivalent async flag in your tool surface). Capture the returned terminal/shell id as $METRO_TERMINAL_ID.

After launch, wait ≤8s for the "Metro waiting on" line, then:

1. Read the terminal output once (BashOutput with the captured id).
2. Extract the native Metro URL from the terminal output:
   • Locate the line beginning › Metro: — it has the form exp+<scheme>://expo-development-client/?url=<encoded-http-url>. Capture the full Metro URL.
3. Generate QR code PNG and present it to the user (chat-first, deterministic fallback):

• Run npx --yes qrcode -o <working_dir>/.expo/metro-qr.png "<metro-url>" to generate the PNG. If the project's npm config requires auth and the fetch fails with E401, retry once with npm_config_registry=https://registry.npmjs.org/ npm_config_always_auth=false prefixed.
• Verify the PNG was created: test -f <working_dir>/.expo/metro-qr.png (exit code 0 = success). If it fails, print the qrcode error and continue to step 4.
• Chat-first render (best effort): read and base64-encode the file (base64 <working_dir>/.expo/metro-qr.png) and embed in markdown as a data URI (![QR](data:image/png;base64,<data>)) so hosts that support inline image markdown show the QR directly in chat.
• Guaranteed visible fallback: if inline chat image rendering is unavailable in the host UI, open the PNG directly in the default system image viewer/browser (open <working_dir>/.expo/metro-qr.png on macOS, xdg-open ... on Linux, start "" ... on Windows). This fallback is required whenever chat image rendering is unavailable.
• Surface only the native Metro URL immediately after the image/fallback message.

4. Optional: ASCII terminal QR for power users. Extract and print the terminal's ASCII QR banner as a secondary/backup option:
   • Locate the first line composed of unicode block glyphs (▀ ▄ █) — that is the top of the QR.

• Print every line from that line through the › Metro: line.
• Cap at 30 lines as a safety net. Print as-is inside a fenced code block so terminal renderers preserve glyph alignment.
• If the ASCII QR banner is not yet in the output, re-read BashOutput once more after another 4s before giving up. If still absent, skip the ASCII QR — PNG delivery from step 3 is the primary path.

5. Follow with:

   "✓ Metro is running in background terminal <id>.

📱 Scan the QR code shown above (or opened from <working_dir>/.expo/metro-qr.png) with your native dev client to load the app. Metro URL: <metro-url> 🔄 Edits hot-reload automatically."

Persist the terminal id to memory bank so resumed sessions and downstream skills (/preview-screens, /edit-app, /add-*) can find it:

## Project facts
...
- Metro terminal id: <id> (started <ISO date>)
- Metro launch cmd: cd <working_dir> && npx expo start

This skill stops after Step 12 so the user can iterate locally. Production build + tenant push is a separate, explicit user action via the /deploy skill.

Step 12.5 — Optional debug handoff

Do not perform screen-by-screen runtime verification. Do not crawl routes, open browser targets, use React Native Web, or call Metro HTTP endpoints directly.

After Metro is running and the QR has been presented, offer a single optional debug handoff:

"If the app shows an error or a workflow looks wrong after you load it in the native dev client, tell me the symptom and I can run /debug-app "<symptom>" using the Metro terminal logs."

Only invoke /debug-app if the user asks for debugging or gives a concrete symptom. /debug-app must use the captured Metro terminal output as its diagnostic source; it must not probe localhost, request a bundle URL, or run any React Native Web setup. If the user gives no symptom, proceed directly to Step 13.

When the user is ready to deploy:

/deploy            # runs npm run build + npx power-apps push

Step 13 — Summary

Print a compact status block, then present exactly 4 options with no explanation. Do not add prose, tips, or "you might want to" text — keep it concise.

✅ Native code app created
─────────────────────────────────────────────
App name      : <displayName>
Project       : <working_dir>
Environment   : <env name> (<env id>)
Data model    : <N tables — M reuse, K extend, L create>
Native caps   : <list>
Connectors    : <list>
Screens       : <N total — M from template, K built in parallel>
Dev server    : npx expo start — running in background terminal <id>
                (scan QR there when you want to run locally)
─────────────────────────────────────────────

If Step 1 emitted warnings, list them in one line each under the block (no decoration).

Then present exactly these 4 options:

What now?

1. Preview screens in browser  (/preview-screens)
2. Deploy to tenant            (/deploy)
3. Edit the app                (/edit-app)
4. Add more capabilities       (/add-dataverse, /add-connector, /add-native)
5. Configure auth later        (/set-app-registration-native)

Which option? (or "none — I'll keep iterating locally")

Hard rules for this step:

• Do NOT add explanatory paragraphs after the options.
• Do NOT recommend an option ("most users want #2").
• Do NOT list alternative npm commands — the dev server is already running and is the only local iteration process the user needs to know about.
• Wait for the user's choice before doing anything else. If they pick none, stop.

Notes

• This skill is the only entry point for new project creation. Do not invoke /add-* skills directly during a fresh-project flow — they don't know how to read the plan and would re-prompt the user.
• The planner agent owns the approval gates. This skill never enters plan mode itself — that would create a duplicate gate.
• For mid-project changes after Step 13, the user should run individual /add-* skills, or /edit-app for plan-backed app iteration.

Reference

• shared/shared-instructions.md
• shared/references/screen-templates.md
• agents/native-app-planner.md