qa-find-bugs-mobile

Hunt bugs in the Beancount iOS app using Expo MCP on the iPhone 17e simulator.

275|32|Updated Oct 2, 2020
One-click install
npx skills add https://github.com/bex-co/beancount-io --skill qa-find-bugs-mobile-bex-co
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qa-find-bugs-mobile
Source: https://github.com/bex-co/beancount-io/tree/main/.agents/skills/qa-find-bugs-mobile
Command: npx skills add https://github.com/bex-co/beancount-io --skill qa-find-bugs-mobile-bex-co

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manually testing a native mobile app across every user journey is slow and error-prone, and browser-based mobile emulation misses native-only defects. This Skill drives the real Beancount iOS app on the iPhone 17e simulator through Expo MCP, reproducing failures, tracing root causes, and deduplicating findings. ## Core Features & Use Cases - Native device automation: Boots the exact iPhone 17e simulator, launches the io.beancount.ios development build, and drives it via the local Expo MCP server with tap, view inspection, screenshot, and log collection tools. - Real authentication flows: Loads QA_EMAIL and QA_PASSWORD from mobile/.env, completes the native OAuth/PKCE sign-in, and verifies session persistence across relaunches without fabricating tokens. - Full journey sweeps: Systematically exercises auth, discovery, home, transactions, entry, accounts, reports, merchants, files, settings, and accessibility journeys with evidence capture and triage. - Use Case: Before a mobile release, run the skill to sweep all journeys on iPhone 17e, reproduce a transaction-filter bug, trace it to mobile/src/screens code, and file a deduplicated finding. ## Quick Start Run the qa-find-bugs-mobile skill to test the Beancount app on the iPhone 17e simulator and report any bugs you find.

Frequently Asked Questions about qa-find-bugs-mobile

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate QA testing of an Expo iOS app on a simulator?

Boot the target simulator, launch the native development build, and start the local Expo MCP server with node node_modules/expo-mcp/bin/expo-mcp.mjs pointing at the Metro dev server. Then drive the app with automation_tap, automation_find_view, automation_take_screenshot, and collect_app_logs over JSON-RPC stdio.

How do I sign in to a mobile app during automated QA testing?

Load QA_EMAIL and QA_PASSWORD from mobile/.env inside a credential helper, tap the native Sign In action, and complete the browser OAuth/PKCE authorization. Verify control returns to the app and the session survives a relaunch; never inject a fabricated token into reactive variables.

Can Expo MCP control multiple booted iOS simulators at once?

No. The pinned Expo MCP version chooses the sole booted iOS simulator and rejects multiple devices. Record other booted simulators, shut them down temporarily for the QA session, and restore them afterward.

Why are my Expo MCP screenshot coordinates wrong when tapping?

Expo MCP resizes screenshots, so dividing image coordinates by a fixed scale factor produces wrong tap targets. Determine the native dimensions and capture scale, or use inspected view bounds and testID-based taps in logical points instead.

What are the limitations of simulator-based mobile QA?

Simulator testing cannot verify haptics, camera hardware, or Android behavior, and the pinned MCP lacks typing and scrolling tools. Deep links can reach screens but only prove the destination, not the navigation journey, and stale native builds must not be counted as product regressions.