reis-mobile 0.3.5 → 0.4.0

package/README.md

@@ -6,7 +6,7 @@ A mobile app is not a generic project. A code review that does not know about `B
 
 <p align="center">
   <a href="https://github.com/wrsilva/reis-mobile/actions/workflows/ci.yml"><img src="https://github.com/wrsilva/reis-mobile/actions/workflows/ci.yml/badge.svg" alt="CI"></a>
-  <img src="https://img.shields.io/badge/Version-0.3.5-blue" alt="Version 0.3.5">
+  <img src="https://img.shields.io/badge/Version-0.4.0-blue" alt="Version 0.4.0">
   <img src="https://img.shields.io/badge/Claude_Code-plugin-blueviolet" alt="Claude Code plugin">
   <img src="https://img.shields.io/badge/Node.js-22+-339933?logo=node.js&logoColor=white" alt="Node.js 22+">
   <img src="https://img.shields.io/badge/Dependencies-0-brightgreen" alt="Zero dependencies">
@@ -23,9 +23,9 @@ A mobile app is not a generic project. A code review that does not know about `B
 
 🔎 **Detects the stack on its own.** Flutter (app or plugin), React Native (including Expo), Kotlin Multiplatform, native Android and iOS, with languages and target platforms. A Flutter app with `android/` and `ios/` is still Flutter.
 
-👥 **7 agents and 67 mobile skills.** Flutter architect, performance and test engineers, Flutter/Android/iOS staff engineer, native plugin specialist, code reviewer and a lead that coordinates them all, plus skills for BLoC, Riverpod, Firebase, testing, layout, plugins and security.
+👥 **16 agents and 74 mobile skills.** Architect, performance and test engineers for Flutter, native Android, native iOS and React Native, a staff engineer, a native plugin specialist, a code reviewer and a lead that coordinates them all, plus skills for Gradle, Xcode and CocoaPods builds, Jetpack Compose, BLoC, Riverpod, Firebase and testing on every platform, plugins and security.
 
-🌐 **Answers in English or Portuguese.** `reis-mobile init pt` or `reis-mobile init en` sets the language of reports, explanations and debates.
+🌐 **Answers in English or Portuguese.** `reis-mobile init eng` or `reis-mobile init pt` sets the language of reports, explanations and debates.
 
 🧭 **Routes to the right specialist.** Describe the problem in English or Portuguese, and the router identifies the intent, the stack and the platform in focus. "The Android build of my Flutter app broke" loads Flutter **and** Android context.
 
@@ -35,24 +35,26 @@ A mobile app is not a generic project. A code review that does not know about `B
 
 🔒 **Secrets never reach the model.** The diff sent for review goes through a layer that masks API keys, tokens, JWTs, private keys and keystore passwords.
 
-🪶 **No dependencies.** `npx reis-mobile init`, zero npm dependencies, no hooks, no external provider.
+🪶 **Two commands to install.** `npm install -g reis-mobile` and `reis-mobile init eng` set it up in Claude Code and Codex, with zero npm dependencies, no hooks and no external provider.
 
 ---
 
 ## What's new
 
-> 🆕 **v0.3.0: the specialists debate.** On an architecture decision, a single specialist gives you the answer of their specialty. `/reis-mobile:debate` puts three of them to defend incompatible positions on the same question — and `lead-mobile` decides, instead of listing pros and cons.
+> 🆕 **v0.4.0: every mobile stack.** Native Android, native iOS and React Native get their own architect, performance and test specialists, and `/reis-mobile:debug` finds the cause of a failing build with the toolchain versions in hand.
 >
 > ```bash
-> /reis-mobile:debate Riverpod or BLoC for this app's state?
-> /reis-mobile:debate --rounds 3 offline-first with Firestore or a local cache?
+> /reis-mobile:debug Execution failed for task ':app:compileDebugKotlin'
+> /reis-mobile:debug pod install could not find compatible versions for Firebase
 > ```
 >
-> Round 1 is blind, so nobody anchors the others. Participants do not vary only in role: they vary in model, because three instances of the same model agree for the same wrong reasons.
+> Tests and Firebase are one skill each — `mobile-test` and `mobile-firebase` — with a guide per platform, so a request loads only what your stack needs.
 
 | Version | Highlights |
 |--------|-----------|
-| **v0.3.5** (current) | Installs through npm (`npx reis-mobile init`) and the Claude Code and Codex plugins. Homebrew and the curl/PowerShell installers are discontinued. |
+| **v0.4.0** (current) | Native Android, iOS and React Native: architect, performance and test agents for each; `/reis-mobile:debug` for Gradle, Xcode, CocoaPods and Flutter failures; `mobile-test` and `mobile-firebase` with a guide per platform; the Android team's skills; `.reis-mobile/config.yaml` for monorepos. Skills now use stack prefixes. |
+| **v0.3.6** | `reis-mobile init eng` installs the plugin in Claude Code and Codex at once, skipping whichever is not installed; `init --uninstall` removes it from both. |
+| **v0.3.5** | Installs through npm and the Claude Code and Codex plugins. |
 | **v0.3.4** | Back to the `reis-mobile` name: `reis-mobile` CLI and `/reis-mobile:*` commands, undoing the `mobile` renames of v0.3.2 and v0.3.3. |
 | **v0.3.1** | Answers in English or Portuguese: `reis-mobile init pt\|en` and `reis-mobile lang` save the language for reports, doctor explanations and debates. Agents, skills, commands and docs are now written in English. |
 | **v0.3.0** | `/reis-mobile:debate` command: three specialists with conflicting priorities debate over two rounds and `lead-mobile` decides. Participants vary in role and model; `--external` adds Codex and Gemini. Brings forward the `council` planned for v0.8.0. |
@@ -64,166 +66,133 @@ A mobile app is not a generic project. A code review that does not know about `B
 
 ## Installation
 
-reis-mobile ships through three channels:
+You need **Node.js 22+** and **[Claude Code](https://claude.com/claude-code)**, the **Codex CLI**, or both.
 
-| Channel | What you get |
-|---|---|
-| **Claude Code plugin** | Everything: 7 agents, 67 skills and the `/reis-mobile:*` commands |
-| **Codex plugin** | The skills, plus the CLI if you install it from npm |
-| **npm** | The `reis-mobile` CLI: `init`, `doctor`, `route`, `review`, `lang`, and the router the commands rely on |
-
-Requirements: **Node.js 22+**. For the plugins, **[Claude Code](https://claude.com/claude-code)** or the **Codex CLI**.
-
-### Claude Code (recommended)
-
-```bash
-npx reis-mobile init pt      # registers the plugin and answers in Portuguese; use "en" for English
-```
-
-Or install the CLI globally and keep it around for `doctor`, `route` and `review`:
+### Install
 
 ```bash
 npm install -g reis-mobile
-reis-mobile init pt
-```
-
-Without Node.js tooling, add the plugin straight from GitHub:
-
-```bash
-claude plugin marketplace add https://github.com/wrsilva/reis-mobile.git
-claude plugin install reis-mobile@reis-mobile
+reis-mobile init eng
 ```
 
-Restart Claude Code and run, in your app folder:
+`init` installs the plugin in Claude Code and Codex, whichever you have, and sets the answer language (`eng` or `pt`):
 
 ```text
-/reis-mobile
-/reis-mobile:doctor
-/reis-mobile:review
+✓ Claude Code installed
+✓ Codex       installed
+✓ Language    en (~/.config/reis-mobile/config.json)
 ```
 
-### Codex
+Restart Claude Code and Codex, open your app folder and try it:
+
+| Tool | Try |
+|---|---|
+| Claude Code | `/reis-mobile:doctor`, then `/reis-mobile:review` |
+| Codex | Type `$` to search the skills, or use `/skills` |
 
-With the Codex CLI installed and supporting `codex plugin`:
+### Update
 
 ```bash
-codex plugin marketplace add https://github.com/wrsilva/reis-mobile.git
-codex plugin add reis-mobile@reis-mobile
+npm update -g reis-mobile
+reis-mobile init eng
 ```
 
-The first command registers the marketplace with GitHub as its source; the second installs the plugin into the Codex cache. Check it with `codex plugin list --json`: look for `reis-mobile@reis-mobile` with `installed: true` and `enabled: true`.
-
-Restart Codex and open a conversation in your mobile app folder. In the CLI or the IDE extension, type `$` in the message field to search for skills, or use `/skills`. See the [official Codex skills documentation](https://learn.chatgpt.com/docs/build-skills#how-chatgpt-and-codex-use-skills).
+Then restart Claude Code and Codex. See what changed in the [CHANGELOG](CHANGELOG.md).
 
-The plugin gives Codex the skills; the agents and the `/reis-mobile:*` commands are Claude Code formats and have not been validated in Codex. For routing and review context, install the CLI from npm and run it in the app folder:
+### Uninstall
 
 ```bash
-npm install -g reis-mobile
-reis-mobile doctor
-reis-mobile route "review architecture" --json
-reis-mobile review --json
+reis-mobile init --uninstall
+npm uninstall -g reis-mobile
 ```
 
-You can also ask Codex: *"Run `reis-mobile review --json`, read the instructions in the returned files and review the changes using that context."* The CLI selects the instructions and gathers the context; Codex does the analysis.
+The first command removes the plugin from Claude Code and Codex and deletes the saved language; the second removes the CLI. Keep the `-g`: without it, npm looks for the package in the current folder and removes nothing.
 
-### Language
+<details>
+<summary>Install without npm</summary>
 
-Agents, skills and commands are written in English, and that does not change. What you choose is the language reis-mobile **answers** in: `/reis-mobile:review` reports, `/reis-mobile:doctor` explanations, and debate rounds and syntheses.
+Add the plugin straight from GitHub. The `marketplace add` line only needs to run once per machine.
 
 ```bash
-reis-mobile init pt          # install the plugin and answer in Brazilian Portuguese
-reis-mobile init en          # install the plugin and answer in English (also: eng)
-reis-mobile lang pt          # change the language later, without reinstalling
-reis-mobile lang             # show the current language
+# Claude Code
+claude plugin marketplace add https://github.com/wrsilva/reis-mobile.git
+claude plugin install reis-mobile@reis-mobile
+
+# Codex
+codex plugin marketplace add https://github.com/wrsilva/reis-mobile.git
+codex plugin add reis-mobile@reis-mobile
 ```
 
-The choice is saved to `~/.config/reis-mobile/config.json` (`%APPDATA%\reis-mobile\config.json` on Windows), so it survives `claude plugin update`. Precedence is `--lang` on a single run, then the `REIS_MOBILE_LANG` environment variable, then the saved file. With nothing set, each command answers in the language of your request. Agents that Claude Code triggers on its own, outside a `/reis-mobile:*` command, follow the language of the conversation.
+To update, run `claude plugin marketplace update reis-mobile` and `claude plugin update reis-mobile@reis-mobile` in Claude Code, or `codex plugin marketplace upgrade reis-mobile` and `codex plugin add reis-mobile@reis-mobile` in Codex.
 
-<details>
-<summary>Uninstall</summary>
+To uninstall:
 
 ```bash
-reis-mobile init --uninstall        # Claude Code plugin, its marketplace and the saved language
-npm uninstall -g reis-mobile        # CLI
+# Claude Code
+claude plugin uninstall reis-mobile@reis-mobile
+claude plugin marketplace remove reis-mobile
 
+# Codex
 codex plugin remove reis-mobile@reis-mobile
 codex plugin marketplace remove reis-mobile
 ```
 </details>
 
 <details>
-<summary>Installed with Homebrew or the curl/PowerShell installer?</summary>
+<summary>Language</summary>
 
-Those channels were discontinued after v0.3.4. Remove them and install from npm:
+Agents, skills and commands are written in English. The language you choose is the one reis-mobile **answers** in: `/reis-mobile:review` reports, `/reis-mobile:doctor` explanations, and debate rounds and syntheses.
 
 ```bash
-# Homebrew
-brew uninstall reis-mobile && brew untap wrsilva/tap
-
-# curl installer (macOS/Linux)
-rm -rf ~/.local/share/reis-mobile ~/.local/bin/reis-mobile
+reis-mobile lang pt          # switch to Brazilian Portuguese without reinstalling
+reis-mobile lang eng         # switch to English
+reis-mobile lang             # show the current language
 ```
 
-```powershell
-# PowerShell installer (Windows)
-Remove-Item -Recurse -Force "$env:LOCALAPPDATA\reis-mobile"
-Remove-Item -Force "$HOME\.local\bin\reis-mobile.cmd"
-```
+The choice is saved to `~/.config/reis-mobile/config.json` (`%APPDATA%\reis-mobile\config.json` on Windows) and survives plugin updates. `--lang` on a single run and the `REIS_MOBILE_LANG` environment variable override it. With nothing set, each command answers in the language of your request. Agents that Claude Code triggers on its own, outside a `/reis-mobile:*` command, follow the language of the conversation.
 </details>
 
 <details>
-<summary>Development from the clone</summary>
+<summary>Using the CLI from Codex</summary>
+
+The Codex plugin gives Codex the skills; the agents and the `/reis-mobile:*` commands are Claude Code formats and have not been validated in Codex. For routing and review context, run the CLI in the app folder:
 
 ```bash
-git clone https://github.com/wrsilva/reis-mobile.git
-cd reis-mobile
-npm link                      # CLI pointing at the clone
-reis-mobile init --local      # plugin pointing at the clone
-npm run check                 # validates and runs the tests
+reis-mobile doctor
+reis-mobile route "review architecture" --json
+reis-mobile review --json
 ```
 
-After editing agents, skills or commands, restart the Claude Code session.
+You can also ask Codex: *"Run `reis-mobile review --json`, read the instructions in the returned files and review the changes using that context."* The CLI selects the instructions and gathers the context; Codex does the analysis.
 </details>
 
----
-
-## Updating
-
-To see what changed, check the [CHANGELOG](CHANGELOG.md) or the [releases](https://github.com/wrsilva/reis-mobile/releases).
-
-```bash
-# Claude Code plugin
-claude plugin marketplace update reis-mobile
-claude plugin update reis-mobile@reis-mobile
-
-# Codex plugin
-codex plugin marketplace upgrade reis-mobile
+<details>
+<summary>Troubleshooting</summary>
 
-# CLI
-npm update -g reis-mobile
-```
+| Symptom | Solution |
+|---------|---------|
+| Codex says `plugin reis-mobile was not found in marketplace reis-mobile` | The marketplace is not registered or is outdated. Run `reis-mobile init eng`, or `codex plugin marketplace add https://github.com/wrsilva/reis-mobile.git` followed by `codex plugin add reis-mobile@reis-mobile` |
+| `claude plugin update` says it is already on the latest version, but the release is newer | Run `claude plugin marketplace update reis-mobile` before `update` |
+| New agents or skills do not show up | Restart Claude Code or Codex: plugins only reload in a new session |
+| `reis-mobile --version` still shows the old version after updating | Another `reis-mobile` comes first on the `PATH`; find it with `which -a reis-mobile` (`where.exe reis-mobile` on Windows) and remove it |
+| `npm uninstall reis-mobile` says `up to date` and the command is still there | Add `-g`: `npm uninstall -g reis-mobile` |
+| `/reis-mobile:*` commands missing after upgrading from v0.3.2 or v0.3.3 | Those versions named the plugin `mobile`. Run `reis-mobile init eng`: it installs `reis-mobile@reis-mobile` and removes `mobile@reis-mobile` |
+| Plugin broken after updating | Reinstall: `reis-mobile init --uninstall`, then `reis-mobile init eng` |
 
-Then **restart Claude Code or Codex**. The plugin update only happens when a new version is out; if you are already on the latest, Claude Code answers `already at the latest version`.
+</details>
 
-Check that the CLI and the plugin match:
+<details>
+<summary>Development from the clone</summary>
 
 ```bash
-reis-mobile --version
-claude plugin list | grep -A1 reis-mobile
+git clone https://github.com/wrsilva/reis-mobile.git
+cd reis-mobile
+npm link                      # CLI pointing at the clone
+reis-mobile init --local      # plugin pointing at the clone
+npm run check                 # validates and runs the tests
 ```
 
-> **Installed v0.3.2 or v0.3.3?** Those versions renamed the plugin and the CLI to `mobile`; v0.3.4 goes back to `reis-mobile`. Run `reis-mobile init`: it installs `reis-mobile@reis-mobile` and removes `mobile@reis-mobile`. In Codex: `codex plugin add reis-mobile@reis-mobile` and `codex plugin remove mobile@reis-mobile`.
-
-<details>
-<summary>Update troubleshooting</summary>
-
-| Symptom | Solution |
-|---------|---------|
-| `claude plugin update` says it is already on the latest version, but the release is newer | Run `claude plugin marketplace update reis-mobile` before `update` |
-| New agents or skills do not show up | Restart Claude Code: plugins only reload in a new session |
-| `reis-mobile --version` is still old after `npm update -g` | Another `reis-mobile` comes first on the `PATH`, often from Homebrew or the old installers; check with `which -a reis-mobile` (`where.exe reis-mobile` on Windows) |
-| Plugin broken after updating | Reinstall: `reis-mobile init --uninstall` and then `reis-mobile init` |
-
+After editing agents, skills or commands, restart Claude Code or Codex.
 </details>
 
 ---
@@ -238,10 +207,17 @@ claude plugin list | grep -A1 reis-mobile
 /reis-mobile:review                               # Reviews uncommitted changes (or the whole project, if there are none)
 /reis-mobile:review --base main                   # Reviews the current branch against main, pull request style
 /reis-mobile:review --base main focus on security # Free-form focus, in English or Portuguese
+/reis-mobile:debug pod install fails on CI        # Finds the cause of a build or runtime failure and proposes the fix
 /reis-mobile:debate Riverpod or BLoC in this app? # Debate between the specialists, with a decision by lead-mobile
 /reis-mobile:debate --rounds 3 --external ...     # More rounds; --external adds Codex and Gemini, if installed
 ```
 
+### Debug
+
+`/reis-mobile:debug` takes the error, the failing command or a description of the problem. It routes to the debug agent and to the skills for the stack and the build area it recognizes (`gradle`, `xcode`, `cocoapods`, `signing`, `pub`...), adds the doctor report with the installed toolchain versions, and answers with the cause, the evidence, the minimal fix and how to verify it.
+
+It proposes; it does not change your project on its own. Commands that alter files or caches — `flutter clean`, `pod install`, deleting `DerivedData` or Gradle caches, dependency upgrades — only run after you ask it to apply the fix.
+
 ### Debate
 
 `/reis-mobile:debate` puts three specialists with conflicting priorities to defend positions on the same question. Round 1 is blind, so the first to answer does not anchor the others; in round 2 each one rebuts specific points from the others; at the end, `lead-mobile` decides and delivers the action plan. The rounds are stored in `.reis-mobile/debates/`.
@@ -259,13 +235,13 @@ $ reis-mobile route "Execution failed for task ':app:compileDebugKotlin'"
 Intent      debug (confidence 0.5) · area gradle
 Stack       flutter · focus android
 Agent       mobile-staff-engineer
-Skills      dart-fix-runtime-errors, dart-resolve-package-conflicts, flutter-errors, flutter-fix-layout-issues
+Skills      flutter-build-debug, flutter-dart-fix-runtime-errors, flutter-dart-resolve-package-conflicts, flutter-errors, flutter-fix-layout-issues, android-gradle-build-debug
 
 $ reis-mobile route "write tests for the login cubit"
 Intent      test (confidence 1)
 Stack       flutter
 Agent       flutter-test-engineer
-Skills      dart-add-unit-test, dart-collect-coverage, dart-generate-test-mocks, flutter-add-integration-test, flutter-add-widget-test, mockito, mocktail, patrol-e2e-testing, testing
+Skills      mobile-test
 ```
 
 The router never invents a specialist: if no agent serves the intent, it warns you. You do not need to call agents by name, because Claude Code triggers them by their description when the request fits. To force one of them, ask: *"use reis-mobile:flutter-architect to review the architecture"*.
@@ -280,15 +256,14 @@ The router never invents a specialist: if no agent serves the intent, it warns y
 | Review a PR or my changes | `/reis-mobile:review` | ✅ |
 | Audit an entire Flutter project | `/reis-mobile:review` with no pending changes | ✅ |
 | Find out a project's stack | `reis-mobile detect` | ✅ |
-| Review the architecture of a Flutter app | agent `flutter-architect` | ✅ |
-| Find the cause of jank, rebuilds or leaks | agent `flutter-performance-engineer` | ✅ |
-| Write or audit Flutter tests | agent `flutter-test-engineer` | ✅ |
-| Debug a Gradle, Xcode or CocoaPods build | agent `mobile-staff-engineer` | ✅ |
+| Review the architecture of an app | agent `flutter-architect`, `android-architect`, `ios-architect` or `rn-architect` | ✅ |
+| Find the cause of jank, rebuilds, hangs or leaks | agent `flutter-`, `android-`, `ios-` or `rn-performance-engineer` | ✅ |
+| Write or audit tests | agent `flutter-`, `android-`, `ios-` or `rn-test-engineer` | ✅ |
+| Debug a Gradle, Xcode, CocoaPods or Flutter build | `/reis-mobile:debug` | ✅ |
 | Build a plugin or debug MethodChannel/EventChannel | agent `plugin-native-expert` | ✅ |
 | Full audit with several specialists | agent `lead-mobile` | ✅ |
 | Decide between two architectures with a real trade-off | `/reis-mobile:debate` | ✅ |
-| Dedicated debug command | `/reis-mobile:debug` | 🔜 v0.4 |
-| Dedicated test command (including XCTest and Espresso) | `/reis-mobile:test` | 🔜 v0.4 |
+| Dedicated test command (including XCTest and Espresso) | `/reis-mobile:test` | 🔜 v0.5 |
 | Check whether the app is ready for the store | `/reis-mobile:release` | 🔜 v0.5 |
 | Consolidated decision by consensus between agents | `/reis-mobile:council` | 🔜 v0.8 |
 
@@ -301,8 +276,8 @@ The router never invents a specialist: if no agent serves the intent, it warns y
 | **Project context** | You explain the stack | Deterministic detection of stack, platform and variant |
 | **Review** | Generic | Mobile checklists: lifecycle, `BuildContext`, MASVS, manifest, ATS |
 | **Secrets in the diff** | Sent as they are | Masked before reaching the model |
-| **Specialists** | None | 7 agents and 67 mobile skills |
-| **Context cost** | Zero | ~4,800 fixed tokens (agent and skill descriptions), no hooks |
+| **Specialists** | None | 16 agents and 74 mobile skills |
+| **Context cost** | Zero | ~8,400 fixed tokens (agent and skill descriptions), no hooks |
 | **Best for** | General tasks | Teams and developers working on mobile apps |
 
 **In short:** Claude Code already knows how to code. reis-mobile makes it look at what matters in a mobile app.
@@ -319,7 +294,7 @@ The router never invents a specialist: if no agent serves the intent, it warns y
      ├─ detect stack ──────── pubspec.yaml → flutter (android, ios)
      ├─ detect intent ─────── review
      ├─ select agent ──────── mobile-code-reviewer
-     ├─ select skills ─────── code-review · effective-dart · flutter-project-audit · flutter-widget-review · mobile-security-audit …
+     ├─ select skills ─────── flutter-code-review · flutter-effective-dart · flutter-project-audit · flutter-widget-review · mobile-security-audit …
      ├─ collect context ───── git diff (lock files out, secrets masked)
      └─ review report ─────── Summary · Critical · Bugs · Architecture · Security · Performance · Maintainability
 ```
@@ -344,28 +319,37 @@ The Node.js core decides **what** to load. The model decides **how** to review,
 | `flutter-architect` | Flutter | architecture | Layers, feature-first, coupling, misplaced logic, refactoring plan |
 | `flutter-performance-engineer` | Flutter | performance | Rebuilds, jank, lists, leaks, paint, startup, with a score and top 3 fixes |
 | `flutter-test-engineer` | Flutter | test | Unit, BLoC/Cubit, widget and integration, coverage audit and fragile tests |
+| `android-architect` | Android | architecture | UI, domain and data layers, Gradle modules, ViewModel and DI boundaries |
+| `android-performance-engineer` | Android | performance | Compose recomposition, lists, ANRs, startup, leaks, app size |
+| `android-test-engineer` | Android | test | ViewModel and coroutine tests, Compose and Espresso UI tests, flaky tests |
+| `ios-architect` | iOS | architecture | SwiftUI and UIKit structure, packages, navigation, concurrency boundaries |
+| `ios-performance-engineer` | iOS | performance | Hangs, SwiftUI updates, lists, launch time, retain cycles, memory |
+| `ios-test-engineer` | iOS | test | Swift Testing and XCTest unit tests, async code, XCUITest flows |
+| `rn-architect` | React Native | architecture | Feature structure, server vs client state, navigation, native modules |
+| `rn-performance-engineer` | React Native | performance | Re-renders, lists, JavaScript thread, animations, startup, bundle size |
+| `rn-test-engineer` | React Native | test | Jest, React Native Testing Library, native module mocks, Detox or Maestro |
 | `mobile-staff-engineer` | all | debug, architecture, performance, test, security, release, migration, dependency, build, deployment, accessibility | Senior Flutter/Android/iOS specialist for whatever no specific agent covers |
 | `plugin-native-expert` | Flutter | on demand | MethodChannel, EventChannel, Pigeon and Kotlin/Swift bridges, threading and lifecycle |
 | `lead-mobile` | all | on demand | Coordinates the other agents and consolidates a single action plan |
 
-When more than one agent serves the intent, the stack-specific one wins. In a Flutter app, `performance` goes to `flutter-performance-engineer`; in a native Android app, it goes to `mobile-staff-engineer`.
+When more than one agent serves the intent, the stack-specific one wins. In a Flutter app, `performance` goes to `flutter-performance-engineer`; in a native Android app, to `android-performance-engineer`. Intents without a stack specialist, such as `debug` or `release`, go to `mobile-staff-engineer`.
 
 ### Skills
 
-**Auto-routed** skills are loaded by the router together with the agent. The others (—) are triggered by Claude Code when the request matches the description, for example *"add push notifications with FCM"* → `firebase-messaging`.
+**Auto-routed** skills are loaded by the router together with the agent. The others (—) are triggered by Claude Code when the request matches the description, for example *"add push notifications with FCM"* → `mobile-firebase`.
 
 <details>
 <summary><strong>Review and quality</strong> (7)</summary>
 
 | Skill | What it does | Auto-routed for | Stack | Origin |
 |---|---|---|---|---|
-| `code-review` | Performs thorough code reviews for Flutter/Dart pull requests and merge requests. | review | Flutter | evanca/flutter-ai-rules |
-| `effective-dart` | Applies Effective Dart guidelines in Flutter/Dart code. | review | Flutter | evanca/flutter-ai-rules |
-| `dart-run-static-analysis` | Execute `dart analyze` to identify warnings and errors, and use `dart fix --apply` to automatically resolve… | review | Flutter | dart-lang/skills |
+| `flutter-code-review` | Performs thorough code reviews for Flutter/Dart pull requests and merge requests. | review | Flutter | evanca/flutter-ai-rules |
+| `flutter-effective-dart` | Applies Effective Dart guidelines in Flutter/Dart code. | review | Flutter | evanca/flutter-ai-rules |
+| `flutter-dart-run-static-analysis` | Execute `dart analyze` to identify warnings and errors, and use `dart fix --apply` to automatically resolve… | review | Flutter | dart-lang/skills |
 | `flutter-project-audit` | Audits the health of a Flutter project as a whole — pubspec constraints and lock file, analysis_options and lints,… | review, architecture, dependency | Flutter | reis-mobile |
 | `flutter-widget-review` | Reviews Flutter widget code for lifecycle bugs, BuildContext misuse across async gaps, missing dispose, side effects… | review, performance | Flutter | reis-mobile |
-| `dart-3-updates` | Applies Dart 3 language features in Flutter/Dart code. | migration | Flutter | evanca/flutter-ai-rules |
-| `dart-use-pattern-matching` | Applies Dart 3 pattern matching, switch expressions, and destructuring idiomatically to validate data schemas,… | — | Flutter | dart-lang/skills |
+| `flutter-dart-3-updates` | Applies Dart 3 language features in Flutter/Dart code. | migration | Flutter | evanca/flutter-ai-rules |
+| `flutter-dart-use-pattern-matching` | Applies Dart 3 pattern matching, switch expressions, and destructuring idiomatically to validate data schemas,… | — | Flutter | dart-lang/skills |
 
 </details>
 
@@ -374,45 +358,37 @@ When more than one agent serves the intent, the stack-specific one wins. In a Fl
 
 | Skill | What it does | Auto-routed for | Stack | Origin |
 |---|---|---|---|---|
-| `architecture-feature-first` | Structures Flutter apps using layered architecture (UI / Logic / Data) with feature-first file organization. | architecture | Flutter | evanca/flutter-ai-rules |
+| `flutter-architecture-feature-first` | Structures Flutter apps using layered architecture (UI / Logic / Data) with feature-first file organization. | architecture | Flutter | evanca/flutter-ai-rules |
 | `flutter-app-architecture` | Provides best practices for Flutter app architecture, including layered architecture, data flow, state management… | architecture | Flutter | evanca/flutter-ai-rules |
 | `flutter-apply-architecture-best-practices` | Architects a Flutter application using the recommended layered approach (UI, Logic, Data). | architecture | Flutter | flutter/skills |
 | `flutter-managing-state` | Manages application and ephemeral state in a Flutter app. | architecture | Flutter | flutter/skills |
-| `bloc` | Implements Flutter state management using the bloc library (Bloc and Cubit). | — | Flutter | evanca/flutter-ai-rules |
-| `riverpod` | Uses Riverpod for state management in Flutter/Dart. | — | Flutter | evanca/flutter-ai-rules |
-| `provider` | Uses the Provider package for dependency injection and state management in Flutter. | — | Flutter | evanca/flutter-ai-rules |
+| `flutter-bloc` | Implements Flutter state management using the bloc library (Bloc and Cubit). | — | Flutter | evanca/flutter-ai-rules |
+| `flutter-riverpod` | Uses Riverpod for state management in Flutter/Dart. | — | Flutter | evanca/flutter-ai-rules |
+| `flutter-provider` | Uses the Provider package for dependency injection and state management in Flutter. | — | Flutter | evanca/flutter-ai-rules |
 | `flutter-change-notifier` | Implements state management with ChangeNotifier and Provider in Flutter. | — | Flutter | evanca/flutter-ai-rules |
 | `flutter-login-usecase` | Implements a Flutter login use case that authenticates through an injected repository and persists the access token… | — | Flutter | reis-mobile |
 
 </details>
 
 <details>
-<summary><strong>Testing</strong> (10)</summary>
+<summary><strong>Testing</strong> (1)</summary>
 
 | Skill | What it does | Auto-routed for | Stack | Origin |
 |---|---|---|---|---|
-| `testing` | Writes and reviews Flutter/Dart tests. | test | Flutter | evanca/flutter-ai-rules |
-| `dart-add-unit-test` | Write and organize unit tests for functions, methods, and classes using `package:test`. | test | Flutter | dart-lang/skills |
-| `flutter-add-widget-test` | Implement a component-level test using `WidgetTester` to verify UI rendering and user interactions (tapping,… | test | Flutter | flutter/skills |
-| `flutter-add-integration-test` | Configures Flutter Driver for app interaction and converts MCP actions into permanent integration tests. | test | Flutter | flutter/skills |
-| `patrol-e2e-testing` | Generates and maintains end-to-end tests for Flutter apps using Patrol. | test | Flutter | evanca/flutter-ai-rules |
-| `mocktail` | Uses the Mocktail package for mocking in Flutter/Dart tests. | test | Flutter | evanca/flutter-ai-rules |
-| `mockito` | Uses the Mockito package for mocking in Flutter/Dart tests. | test | Flutter | evanca/flutter-ai-rules |
-| `dart-generate-test-mocks` | Define and generate mock objects for external dependencies using `package:mockito` and `build_runner`. | test | Flutter | dart-lang/skills |
-| `dart-collect-coverage` | Collect coverage using the coverage packge and create an LCOV report | test | Flutter | dart-lang/skills |
-| `dart-migrate-to-checks-package` | Replace the usage of `expect` and similar functions from `package:matcher` to `package:checks` equivalents. | migration | Flutter | dart-lang/skills |
+| `mobile-test` | Unit, widget, integration and end-to-end tests, mocks and coverage for Flutter, native Android, native iOS and React Native, with one reference per platform. | test | all | reis-mobile, flutter/skills, dart-lang/skills, evanca/flutter-ai-rules, android/skills |
 
 </details>
 
 <details>
-<summary><strong>Debug</strong> (4)</summary>
+<summary><strong>Debug</strong> (5)</summary>
 
 | Skill | What it does | Auto-routed for | Stack | Origin |
 |---|---|---|---|---|
 | `flutter-errors` | Diagnoses and fixes common Flutter errors. | debug | Flutter | evanca/flutter-ai-rules |
 | `flutter-fix-layout-issues` | Fixes Flutter layout errors (overflows, unbounded constraints) using Dart and Flutter MCP tools. | debug | Flutter | flutter/skills |
-| `dart-fix-runtime-errors` | Uses get_runtime_errors and lsp to fetch an active stack trace, locate the failing line, apply a fix, and verify… | debug | Flutter | dart-lang/skills |
-| `dart-resolve-package-conflicts` | Workflow for fixing package version conflicts. | debug, dependency | Flutter | dart-lang/skills |
+| `flutter-dart-fix-runtime-errors` | Uses get_runtime_errors and lsp to fetch an active stack trace, locate the failing line, apply a fix, and verify… | debug | Flutter | dart-lang/skills |
+| `flutter-dart-resolve-package-conflicts` | Workflow for fixing package version conflicts. | debug, dependency | Flutter | dart-lang/skills |
+| `flutter-build-debug` | Diagnoses failing Flutter builds by locating the layer that broke — Dart compilation, pub version solving… | debug, build, dependency, migration | Flutter | reis-mobile |
 
 </details>
 
@@ -466,13 +442,12 @@ When more than one agent serves the intent, the stack-specific one wins. In a Fl
 </details>
 
 <details>
-<summary><strong>Security</strong> (3)</summary>
+<summary><strong>Security</strong> (2)</summary>
 
 | Skill | What it does | Auto-routed for | Stack | Origin |
 |---|---|---|---|---|
 | `mobile-security-audit` | Security audit for mobile apps based on the OWASP MASVS categories — insecure token storage, hardcoded secrets,… | review, security, release | all | reis-mobile |
 | `flutter-secure-token-store` | Stores JWT access tokens on Flutter with platform secure storage (Keychain on iOS, Keystore-backed storage on… | security | Flutter | reis-mobile |
-| `firebase-app-check` | Integrates Firebase App Check into Flutter apps. | security | Flutter | evanca/flutter-ai-rules |
 
 </details>
 
@@ -487,23 +462,55 @@ When more than one agent serves the intent, the stack-specific one wins. In a Fl
 </details>
 
 <details>
-<summary><strong>Firebase</strong> (13)</summary>
+<summary><strong>Firebase</strong> (1)</summary>
 
 | Skill | What it does | Auto-routed for | Stack | Origin |
 |---|---|---|---|---|
-| `flutterfire-configure` | Sets up Firebase for Flutter apps using FlutterFire CLI. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-auth` | Integrates Firebase Authentication into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-cloud-firestore` | Integrates Cloud Firestore into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-database` | Integrates Firebase Realtime Database into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-storage` | Integrates Firebase Cloud Storage into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-cloud-functions` | Calls Firebase Cloud Functions from Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-messaging` | Integrates Firebase Cloud Messaging (FCM) into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-in-app-messaging` | Integrates Firebase In-App Messaging into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-remote-config` | Integrates Firebase Remote Config into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-analytics` | Integrates Firebase Analytics into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-crashlytics` | Integrates Firebase Crashlytics into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-ai` | Integrates Firebase AI Logic into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
-| `firebase-data-connect` | Integrates Firebase Data Connect into Flutter apps. | — | Flutter | evanca/flutter-ai-rules |
+| `mobile-firebase` | Firebase on Flutter, native Android, native iOS and React Native: setup, Authentication, Firestore, Realtime Database, Storage, Functions, Messaging, Analytics, Crashlytics, Remote Config, App Check, In-App Messaging, AI Logic and SQL Connect, with a guide per platform and product. | — | all | reis-mobile, evanca/flutter-ai-rules |
+
+</details>
+
+<details>
+<summary><strong>Native Android</strong> (25)</summary>
+
+| Skill | What it does | Auto-routed for | Stack | Origin |
+|---|---|---|---|---|
+| `android-code-review` | Reviews native Android code in Kotlin and Java for the bugs that reach production — lifecycle and Context leaks… | review, performance | Android, kotlin-multiplatform | reis-mobile |
+| `android-gradle-build-debug` | Diagnoses failing Android Gradle builds — JDK, Gradle, Android Gradle Plugin and Kotlin version mismatches… | debug, build, dependency, migration | Android, kotlin-multiplatform | reis-mobile |
+| `android-intent-security` | Best practices for Android Intent security. | security, review | Android | android/skills |
+| `android-agp-9-upgrade` | Upgrades, or migrates, an Android project to use Android Gradle Plugin (AGP) version 9. | migration, build, dependency | Android | android/skills |
+| `android-r8-analyzer` | Analyzes Android build files and R8 keep rules to identify redundancies, broad package-wide rules, and rules that… | performance, build | Android | android/skills |
+| `android-profiler` | Manages Android performance profiling and debugging. | performance | Android | android/skills |
+| `android-play-policy-insights` | Automated auditor designed to verify Android applications against Google Play Policy domains. | release | Android | android/skills |
+| `android-play-billing-library-version-upgrade` | Use this skill when upgrading or migrating an Android project from any legacy Google Play Billing Library (PBL)… | — | Android | android/skills |
+| `android-engage-sdk-integration` | Helps developers integrate, debug, and resolve Play Engage SDK implementation issues. | — | Android | android/skills |
+| `android-migrate-xml-views-to-jetpack-compose` | Provides a structured workflow for migrating an Android XML View to Jetpack Compose. | — | Android | android/skills |
+| `android-adaptive` | Instructions to make or update an app's UI so that it adapts to different Android devices including phones… | — | Android | android/skills |
+| `android-edge-to-edge` | Use this skill to migrate your Jetpack Compose app to add adaptive edge-to-edge support and troubleshoot common… | — | Android | android/skills |
+| `android-styles` | Use this skill to integrate the Jetpack Compose Styles API into an Android project. | — | Android | android/skills |
+| `android-navigation-3` | Learn how to install and migrate to Jetpack Navigation 3, and how to implement features and patterns such as deep… | — | Android | android/skills |
+| `android-navigation-event` | Intercept back gestures and run Predictive Back animations using the NavigationEvent (androidx.navigationevent)… | — | Android | android/skills |
+| `android-camerax` | Provide technical guidance for Android camera development with CameraX. | — | Android | android/skills |
+| `android-media3-cast-integration` | Implements Google Cast support in Android apps using Jetpack Media3. | — | Android | android/skills |
+| `android-restore-credentials` | Provides knowledge and workflows to implement Android's Restore Credentials feature using the androidx.credentials… | — | Android | android/skills |
+| `android-verified-email` | Provides a complete workflow for implementing verified email retrieval on Android Credential Manager API. | — | Android | android/skills |
+| `android-appfunctions` | Analyzes Android apps to identify key user workflows for AppFunctions such as creating a note, playing media, or… | — | Android | android/skills |
+| `android-ml-kit-genai-prompt-api` | Analyzes Android codebases to implement ML Kit GenAI Prompt API. | — | Android | android/skills |
+| `android-wear-compose-m3` | Expert guidance for working with Wear OS Compose Material3. | — | Android | android/skills |
+| `android-leanback-to-compose-tv-migration` | Provides instructions and architectural patterns for migrating Android TV applications from legacy Leanback UI… | — | Android | android/skills |
+| `android-display-glasses-with-jetpack-compose-glimmer` | Provides guidelines for developing projected Android XR apps for display glasses using the Jetpack Compose Glimmer… | — | Android | android/skills |
+| `android-cli` | Provides instructions for installing and using the `android` CLI. | — | Android | android/skills |
+
+</details>
+
+<details>
+<summary><strong>Native iOS</strong> (3)</summary>
+
+| Skill | What it does | Auto-routed for | Stack | Origin |
+|---|---|---|---|---|
+| `ios-code-review` | Reviews native iOS code in Swift and Objective-C for the bugs that reach production — retain cycles in closures… | review, performance | iOS | reis-mobile |
+| `ios-xcode-build-debug` | Diagnoses failing iOS builds in Xcode and xcodebuild — code signing and provisioning, deployment target… | debug, build, release, migration | iOS | reis-mobile |
+| `ios-cocoapods-debug` | Diagnoses CocoaPods failures in iOS, Flutter and React Native projects — "could not find compatible versions"… | debug, dependency, build, migration | iOS | reis-mobile |
 
 </details>
 
@@ -513,7 +520,7 @@ When more than one agent serves the intent, the stack-specific one wins. In a Fl
 | Skill | What it does | Auto-routed for | Stack | Origin |
 |---|---|---|---|---|
 | `flutter-setting-up-on-macos` | Sets up a macOS environment for Flutter development. | — | Flutter | flutter/skills |
-| `detect-mobile-stack` | Detects the mobile stack of the current project (Flutter, Android, iOS, React Native, Kotlin Multiplatform), its… | — | all | reis-mobile |
+| `mobile-detect-stack` | Detects the mobile stack of the current project (Flutter, Android, iOS, React Native, Kotlin Multiplatform), its… | — | all | reis-mobile |
 
 </details>
 
@@ -527,17 +534,17 @@ Third-party skills keep their original name and license. See [THIRD_PARTY_NOTICE
 
 **Secrets masked.** The diff goes through [`core/security/redact.mjs`](core/security/redact.mjs) before reaching the model. Lock files and generated code (`*.g.dart`, `*.freezed.dart`, `*.pbxproj`) stay out of the diff. Redaction is a protection layer, not a guarantee. See [SECURITY.md](SECURITY.md).
 
-**No telemetry.** `detect`, `doctor`, `route` and `review` make no network calls. Only installation touches the network, to download the package and register the plugin. The AI analysis uses the model of the session you are working in, such as Claude Code or Codex.
+**No telemetry.** `detect`, `doctor`, `route`, `review` and `debug` make no network calls. Only installation touches the network, to download the package and register the plugin. One imported skill, `android-play-policy-insights`, runs Python scripts that download your app's public Google Play listing when the model uses it. The AI analysis uses the model of the session you are working in, such as Claude Code or Codex.
 
 **Exception: `/reis-mobile:debate --external`.** This flag, and only this flag, sends the debate context — including excerpts of the files mentioned in the question — to the `codex` and `gemini` CLIs, which are third-party and have their own data policies. Without the flag, nothing leaves your session. The `review` diff is masked by `redact.mjs`, but the context you cite in a debate question does not go through that layer: check what you are sending before using `--external`.
 
-**Context cost.** The descriptions of the 7 agents and 67 skills add up to about 4,800 fixed tokens per session (measured with `claude plugin details reis-mobile`). The full content of each skill is only loaded when it is used.
+**Context cost.** The descriptions of the 16 agents and 74 skills add up to about 8,400 fixed tokens per session, estimated from their length (v0.3 measured about 4,800 with `claude plugin details reis-mobile`). The full content of each skill is only loaded when it is used.
 
 **No hooks.** The plugin does not attach to Claude Code events. It only acts when you call a command or when a skill is relevant.
 
 **Own namespace.** Commands live under `/reis-mobile:*` and do not conflict with the built-in `/review` or `/security-review`.
 
-**Clean uninstall.** `reis-mobile init --uninstall` removes the plugin, its marketplace and the saved language without leaving configuration behind.
+**Clean uninstall.** `reis-mobile init --uninstall` removes the plugin and its marketplace from Claude Code and Codex, and the saved language, without leaving configuration behind.
 
 ---
 
@@ -552,7 +559,8 @@ Third-party skills keep their original name and license. See [THIRD_PARTY_NOTICE
 | v0.3.1 | Answer language (`reis-mobile init pt\|en`, `reis-mobile lang`) and project content in English | ✅ |
 | v0.3.4 | Back to the `reis-mobile` name after the `mobile` renames of v0.3.2 and v0.3.3 | ✅ |
 | v0.3.5 | Distribution through npm and the Claude Code and Codex plugins only | ✅ |
-| v0.4.0 | Native Android and iOS skills, `.reis-mobile/config.yaml` and `/reis-mobile:debug` (Gradle, Xcode, CocoaPods, Flutter) | ⏳ |
+| v0.3.6 | `reis-mobile init` installs in Claude Code and Codex | ✅ |
+| v0.4.0 | Native Android and iOS skills, architect, performance and test agents for Android, iOS and React Native, `.reis-mobile/config.yaml` and `/reis-mobile:debug` (Gradle, Xcode, CocoaPods, Flutter) | ✅ |
 | v0.5.0 | `/reis-mobile:test` and native tests (XCTest, Espresso) | ⏳ |
 | v0.6.0 | `/reis-mobile:release` with quality gates | ⏳ |
 | v0.7.0 | React Native pack | ⏳ |
@@ -568,16 +576,22 @@ Third-party skills keep their original name and license. See [THIRD_PARTY_NOTICE
 Not for the review, which only reads code. `/reis-mobile:doctor` shows what is missing in case you want to build.
 
 **My app lives inside a monorepo.**
-Run it from the app folder, or use `--dir apps/mobile` in the CLI. The diff is restricted to that folder.
+Commit a `.reis-mobile/config.yaml` at the repository root pointing at the app:
+
+```yaml
+app: apps/mobile
+```
+
+Every command then analyzes `apps/mobile`, wherever in the repository you run it, and the review diff is restricted to that folder. Without the file, run from the app folder or pass `--dir apps/mobile`.
 
 **Does it work with native Android and iOS?**
-Detection, the doctor, the security skill and the `mobile-code-reviewer`, `mobile-staff-engineer` and `lead-mobile` agents already work. Android- and iOS-specific skills arrive in v0.4.0; the 63 imported skills are for Flutter, Dart and Firebase.
+Yes. Native Android has its own architect, performance and test agents, review and Gradle debugging skills, and the Android team's skills from `android/skills` (Compose, AGP 9, R8, Play policy, Wear, TV and more). Native iOS has its own agents and skills for review, Xcode and CocoaPods. React Native has architect, performance and test agents; its skills are still on the roadmap.
 
 **I already have skills with the same names in `~/.claude/skills`.**
 The plugin's skills live in the `reis-mobile:` namespace and do not conflict, but Claude Code loads both descriptions. To save context, remove the global copies the plugin already covers.
 
 **Does it work in Codex?**
-The marketplace and plugin `0.2.1` can be installed straight from GitHub with `codex plugin`. See [installation and usage in Codex](#codex). Installation does not confirm parity with the Claude Code agents and commands; the `reis-mobile` CLI can also be run by Codex to get diagnostics and context.
+Yes: `reis-mobile init eng` installs the plugin in Codex too. See [installation](#install). Installation does not confirm parity with the Claude Code agents and commands; the `reis-mobile` CLI can also be run by Codex to get diagnostics and context.
 
 **Does it work in Cursor?**
 The Cursor integration has not been validated yet. The MCP server remains on the roadmap for v0.8.0.