ExpensifyOfficial·6 Agent Skills Included

App

Build, test, and debug the Expensify app with AI agents

Guides AI agents through developing, testing, and debugging the Expensify React Native app across web, iOS, and Android. Enforces coding standards, offline-first state patterns, and post-edit checks so changes pass review without repeated manual fixes. Automates device testing, browser verification, Sentry analysis, and video evidence capture for pull requests.

5,017|4,003

↗

npx skills add Expensify/App --all -g -y

Available:

Tells the AI agent how the Expensify app is structured, which sub-skills to invoke for device testing, browser testing, and Sentry analysis, and which lint, typecheck, and compiler checks to run after every code change.

All Skills in This Repository (6)

Pure Emerald Level Indicators

📦 In Repo

Expensify

coding-standards

Enforces React Native coding standards for performance, consistency, and clean component architecture.

agent-device

Drives iOS and Android devices to test, debug, and profile the Expensify HybridApp.

onyx

Implements Onyx state management patterns for optimistic updates and offline-first React Native apps.

measure-telemetry-span

Measures Sentry performance span durations by replaying agent-device flows on iOS simulators or Android emulators.

sentry

Analyze Sentry issues, crashes, spans, and performance metrics via MCP tools.

playwright-app-testing

Run Playwright browser UI tests for the Expensify App.

Official

Intermediate

Frequently Asked Questions

FAQPage Schema

How to install Expensify App skills?▼

Run `npx skills add Expensify/App --all -g -y` in your terminal to install all skills in this suite globally.

What do the Expensify App skills do?▼

They help an AI agent write code that follows Expensify's standards, manage offline-first Onyx state, and test changes on browsers and mobile devices.

Can AI test the Expensify app on real devices?▼

Yes. The agent-device skill drives iOS simulators and Android emulators for interactive testing, debugging, and performance profiling.

How to record video evidence for a pull request?▼

The agent-device-evidence skill reads test steps from a GitHub PR or issue and automatically records MP4 videos of the flows on iOS and Android.

Do these skills work with Claude Code?▼

Yes. All skills follow the standard SKILL.md format and run directly inside Claude Code and compatible AI coding agents.