Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/expensify/app/agent-devicenpx skills add Expensify/App --skill agent-devicegit clone --depth 1 https://github.com/Expensify/AppWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00045 | $0.01362 |
| Opus 5 | $0.00023 | $0.00681 |
| Sonnet 5 | $0.00009 | $0.00272 |
| Haiku 4.5 | $0.00005 | $0.00136 |
Grade A, and why
agent-device scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agent-device
Pre-flight (auto)
These checks evaluate at skill load. If any line shows FAIL, stop and surface the fix before running any device command.
agent-device version: !R=0.18.0; V=$(agent-device --version 2>/dev/null); [ -n "$V" ] && [ "$(printf '%s\n%s\n' "$R" "$V" | sort -V | head -1)" = "$R" ] && echo "OK ($V)" || echo "FAIL (need v$R+, got: ${V:-not installed}). Fix: npm install -g agent-device@latest"
Bundled CLI skills dir: !D="$(npm root -g)/agent-device/skills/agent-device"; test -s "$D/SKILL.md" && echo "OK ($D)" || echo "FAIL (missing $D/SKILL.md). Fix: npm install -g agent-device@latest"
HybridApp mode: !M=$(scripts/is-hybrid-app.sh 2>/dev/null | tail -1); [ "$M" = "true" ] && echo "OK (HybridApp)" || echo "FAIL (got: ${M:-unknown}). This skill only supports the HybridApp build - ensure the Mobile-Expensify submodule is present."
Bring-up
Run this sequence the first time the user asks for device interaction in a session, before any open / snapshot / replay.
1. Platform
If the user prompt names ios or android explicitly, use it. Otherwise ask. Only iOS and Android are supported; reject other platforms.
2. Bundle ID
HybridApp dev builds only (the pre-flight gate enforces this).
| Platform | Bundle ID | Build command from App root |
|---|---|---|
ios |
com.expensify.expensifylite |
npm run ios |
android |
org.me.mobiexpensifyg.dev |
npm run android |
3. Confirm dev build is installed
agent-device apps --platform <p> --json
If the resolved bundle ID is missing from the list, STOP and instruct the developer to run the matching build command from the App repository root. The build script detects HybridApp mode and builds the native app from Mobile-Expensify/.
4. Metro
agent-device metro prepare --public-base-url http://localhost:8081 --port 8081 --kind react-native
What ships with it
19 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- flows/macros/android/complete-onboarding.ad 892 B
- flows/macros/android/send-message.ad 679 B
- flows/macros/android/sign-in.ad 673 B
- flows/macros/complete-onboarding.ad 867 B
- flows/macros/go-back.ad 286 B
- flows/macros/send-message.ad 654 B
- flows/macros/web/complete-onboarding.ad 785 B
- flows/macros/web/send-message.ad 683 B
- flows/macros/web/sign-in.ad 746 B
- flows/README.md 8.9 KB
- flows/tests/create-expense-manual.ad 1.6 KB
- flows/tests/open-create-expense.ad 639 B
- flows/tests/open-report.ad 626 B
- flows/tests/open-search-router.ad 500 B
- flows/tests/scan-receipt-init.ad 586 B
- flows/tests/send-message.ad 820 B
- flows/tests/submit-expense.ad 777 B
- flows/tests/switch-home-to-inbox.ad 472 B
- flows/tests/switch-home-to-reports.ad 456 B
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 125 lines · 45 tokens per session scan A 77083dd45ba2
agent-device is a skill published in the GitHub repository Expensify/App (5,015 stars, last pushed yesterday), licensed MIT. It adds 45 tokens to every session and 1,362 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…