Compare commits

..
Author SHA1 Message Date
Brent Vatne 5d8fd510b9 Run explore:binary on a local simulator, and bring the README up to date
explore:binary --simulator local explores a build on a simulator or emulator
it creates on this Mac, with no EAS login: the same devices capture uses,
recorded for the startup sweep, driven by agent-device with its own daemon or
by a local Argent tool-server. The report opens in the local Viewer; --upload
hosts it.

The README now covers local repack and local simulators, a single-path quick
start that reviews a change or explores a build, targets, and requirements
split by what each path needs. The explore-binary, explore-setup and
readme-refresh skills follow.
2026-10-17 15:45:47 -07:00
Brent Vatne 284ae6a780 Local simulators and emulators, capture targets, faster and sturdier runs (#7)
Capture on this machine (local iOS simulators and Android emulators, with agent-device or Argent) as well as on EAS Simulator; name several capture targets per config (a phone, a foldable closed and open, Android) and review them in parallel with shared builds; repack locally with a local simulator; and make captures faster and harder to break: controller fault containment, an Argent fold recovery, upright captures of an opened iPhone Duo, a settle after each root reset, a route retry after an adb drop, emulator port reservation, a startup sweep of what interrupted captures leave running, and agent-device 0.21.28.
2026-10-17 14:55:21 -07:00
16 changed files with 417 additions and 192 deletions

No files matched your search

+91 -80
View File
@@ -1,6 +1,6 @@
# Expo Explore
Explore maps your app: a screenshot of every screen, connected by the taps between them. On each pull request it captures the screens the PR changes, before and after, on EAS Simulator, and posts a visual review.
Explore maps your app: a screenshot of every screen, connected by the taps between them. On each pull request it captures the screens the PR changes, before and after, on EAS Simulator, and posts a visual review. On your Mac it can also run on a local simulator.
<picture>
<source media="(prefers-color-scheme: dark)" srcset="docs/screenshots/readme-hero-dark.png">
@@ -53,52 +53,74 @@ which look like regressions. For each regression, name the source file to fix.
<details>
<summary><strong>Requirements</strong></summary>
- An iOS app, an Android app, or both. Explore is in v0; Android capture is newer than iOS.
- Expo SDK 57 or newer, with Expo Router (57 or newer) or React Navigation (6 or newer).
- An EAS build profile per platform: `ios.simulator: true` for iOS, `android.buildType: apk` for Android. Its EAS environment cannot have secret-visibility variables.
- EAS Simulator access (in preview, [join the waitlist](https://expo.dev/services/simulators)).
- Node.js 24.12 or newer, and admin access to the GitHub repository.
- To review a change: Expo SDK 57 or newer, with Expo Router (57 or newer) or React Navigation (6 or newer), and an Expo account, since Explore builds the app on EAS. An EAS build profile per platform: `ios.simulator: true` for iOS, `android.buildType: apk` (not a development build) for Android. Its EAS environment cannot have secret-visibility or file variables.
- To explore a build: a simulator build of any app, and a Claude Code or Codex login (or an API key). On a local simulator it needs no Expo account.
- Somewhere to run the app: EAS Simulator (in preview, [join the waitlist](https://expo.dev/services/simulators)), which CI uses; or a [local simulator](#simulator) on your Mac: Xcode 26 or newer for iOS (the default devices are iPhone 17 models) and, for Android, the Android SDK with `ANDROID_HOME` set (emulator, platform-tools, cmdline-tools and a system image; to review a change, also build-tools and a JDK 17).
- Node.js 24.12 or newer. For CI, admin access to the GitHub repository.
</details>
## Quick start
Review a change to your own app, on your machine. You need an Expo SDK 57+ app with Expo Router or React Navigation, Node.js 24.12+, and EAS Simulator access ([join the waitlist](https://expo.dev/services/simulators)). No API keys, no CI.
Paste this into Claude Code or Codex from your app's directory. It reviews a change to your app, before and after, or, if you point it to a build, maps every screen of that build:
From your app's directory:
```text
Show me Expo Explore working on this app, with npx @expo/explore. If I point you to a
build (.app, .tar.gz, simulator .ipa, or .apk), explore it: follow
https://unpkg.com/@expo/explore/skills/explore-binary/SKILL.md. If this is not an Expo
project, ask me for a simulator build and explore that. Otherwise review a change to
this app: follow the "Review a change on this machine" section of
https://unpkg.com/@expo/explore/skills/explore-setup/SKILL.md. If the app cannot build
for the iOS Simulator on EAS yet, set that up (an EAS project, ios.bundleIdentifier, an
"explore" profile with ios.simulator: true) and ask me before you commit it. If I have
no EAS Simulator access, pass --simulator local (it needs Xcode; exploring an .apk needs
the Android SDK instead). If I have no uncommitted changes, change some text near the
top of a screen. If the app needs a sign-in, ask me for a test account the way the skill
describes, never for a password in chat. Open the report, and if I need to log in to
something first, tell me the command.
```
1. **Make the app buildable on EAS, if it is not yet.**
You need Node.js 24.12+. Reviewing a change needs an Expo account and an Expo SDK 57+ app with Expo Router or React Navigation: it is built on EAS and captured on EAS Simulator ([join the waitlist](https://expo.dev/services/simulators)) or, with Xcode, on a simulator on your Mac. Exploring a build needs a Claude Code or Codex login, and EAS Simulator or a simulator on your Mac; on your Mac, no Expo account. No API keys, no CI.
<details>
<summary><strong>Run it yourself</strong></summary>
**Review a change**, from your app's directory:
1. Make the app buildable on EAS, if it is not yet:
```sh
npx eas-cli@latest login
npx eas-cli@latest init
```
Then set `ios.bundleIdentifier` in your app config, add a simulator profile to `eas.json`, and commit. The review builds your last commit as the "before".
Then set `ios.bundleIdentifier` in your app config, add a simulator profile to `eas.json`, and commit. The "before" is where your branch left the default branch: on the default branch, your last commit.
```json
{ "build": { "explore": { "ios": { "simulator": true } } } }
```
2. **Change a screen, then review the change.** Edit some text near the top of a screen (Explore captures the first screenful) and run:
2. Edit some text near the top of a screen (Explore captures the first screenful) and run:
```sh
npx @expo/explore capture
npx @expo/explore capture # on EAS Simulator
npx @expo/explore capture --simulator local # or on a simulator on your Mac
```
Explore builds your last commit and your uncommitted changes on EAS, captures the screens the changes can affect, before and after, on EAS Simulator, and opens the report in your browser. The first run waits for an EAS build; later runs reuse it while your native code is unchanged.
Explore builds your last commit and your uncommitted changes on EAS, captures the screens the changes can affect, before and after, and opens the report in your browser. The first run waits for an EAS build; later runs reuse it while your native code is unchanged. A [local simulator](#simulator) is one Explore creates on your Mac and removes afterwards.
Or paste this into Claude Code or Codex from your app's directory:
**Explore a build:**
```text
Show me an Expo Explore review of a change to this app. Follow the "Review a change on
this machine" section of https://unpkg.com/@expo/explore/skills/explore-setup/SKILL.md.
If the app cannot build for the iOS Simulator on EAS yet, set that up (an EAS project,
ios.bundleIdentifier, an "explore" profile with ios.simulator: true) and ask me before
you commit it. If I have no uncommitted changes, change some text near the top of a
screen. Then run npx @expo/explore capture and open the report.
```sh
npx @expo/explore explore:binary ./my-app.apk # Android APK
npx @expo/explore explore:binary ./my-app.app # iOS simulator build (.app, .tar.gz, or simulator .ipa)
npx @expo/explore explore:binary ./my-app.app --simulator local # on a simulator on your Mac
```
Explore starts an EAS Simulator, installs the build, taps through the app with your Claude Code or Codex login, and prints a link to the hosted report. With `--simulator local` it runs on a simulator or emulator it creates on your Mac, needs no Expo account, and opens the report in your browser. Its [options](#options) are below.
</details>
Every run is kept on your machine for 30 days. `npx @expo/explore reports` lists your local reviews, explorations and uploaded reports with their commits, and reopens or hosts one without capturing again:
<picture>
@@ -115,12 +137,12 @@ npx @expo/explore capture --committed # HEAD alone
npx @expo/explore capture --base v1.3.0 --head v1.4.0 --scope all --upload
```
- It checks out both commits as temporary worktrees, installs their dependencies, captures every configured platform on EAS Simulator (`--platform` picks one), and opens the report in a local Viewer. `--upload` hosts it and prints the link; `report:upload <report.zip>` hosts one you already have.
- It checks out both commits as temporary worktrees, installs their dependencies, captures every configured platform or [target](#targets) on EAS Simulator or a [local simulator](#simulator), and opens the report in a local Viewer. `--target` or `--platform` picks some; they run at once, at most `--parallel N` at a time. `--upload` hosts it for 30 days (`--expires-in` changes that) and prints the link; `report:upload <report.zip>` hosts one you already have.
- Your checkout, index and branches are never touched. Your uncommitted changes, untracked files included, are reviewed as `eas build` would upload them; `--committed`, or `cli.requireCommit: true` in `eas.json`, leaves them out.
- It runs both commits' code on your machine, as you: installing their dependencies, evaluating the app config, and bundling the JavaScript. That is what checking them out and running them would do, so review only commits you would run, and someone else's branch only if you trust it.
- Installs run without credential environment variables, which keeps tokens out of reach of a registry a commit's `.npmrc` names; that is not a sandbox. A registry that needs `${NPM_TOKEN}` in the project's `.npmrc` must be authenticated in your own `~/.npmrc`.
- Captures are cached in `~/.cache/expo-explore`, so the same base is not captured twice (`--no-cache` skips it); a capture unused for 30 days goes, and the cache stays under 2 GB. Ctrl+C stops the run's simulator sessions.
- Like a CI review, it builds on EAS (reusing a matching build) and runs simulator sessions: expect several minutes and EAS usage.
- Captures are cached in `~/.cache/expo-explore`, so the same base is not captured twice (`--no-cache` skips it); a capture unused for 30 days goes, and the cache stays under 2 GB. Ctrl+C stops the run's simulator sessions and local devices; when a run is stopped harder than that, the next local run stops what it left running.
- Like a CI review, it builds on EAS (reusing a matching build) and, unless it captures on a local simulator, runs EAS Simulator sessions: expect several minutes and EAS usage.
- Saved runs take up to 2 GB in all, the oldest going first. `reports add <report.zip>` keeps a report you downloaded, such as a PR's `explore-review` artifact.
</details>
@@ -129,44 +151,22 @@ Explore uses its defaults until you run `npx @expo/explore setup`, which writes
## Map every screen of any app
To map every screen of an app instead, with no config, paste this into Claude Code or Codex from your app's directory. It explores a build you point to, or an Expo project straight from its source:
```text
Show me Expo Explore working on this app. Follow
https://unpkg.com/@expo/explore/skills/explore-binary/SKILL.md and run Explore with
npx @expo/explore. If I point you to a build, explore that. Otherwise, if this is an
Expo project, run it from source without making a new build; if it is not, get a
simulator build of it. If the app needs a sign-in, ask me for a test account the way
the skill describes, never for a password in chat. Give me the report link, and if I
need to log in to something first, tell me the command.
```
For apps behind a sign-in, see [Sign-in and other app state](#sign-in-and-other-app-state).
<details>
<summary><strong>Run it yourself</strong></summary>
```sh
npx @expo/explore explore:binary ./my-app.apk # Android APK
npx @expo/explore explore:binary ./my-app.app # iOS simulator build (.app, .tar.gz, or simulator .ipa)
```
Explore starts an EAS Simulator, installs the build, taps through the app with your Claude Code or Codex login, and prints a link to the hosted report.
</details>
Pointed to a build, the [quick start](#quick-start) maps every screen of it, with no source and no config: one card per screen it reached, connected by the taps between them. For apps behind a sign-in, see [Sign-in and other app state](#sign-in-and-other-app-state).
<details>
<summary id="options"><strong>What it needs, and its options</strong></summary>
- An EAS login (`eas whoami`) and a `claude` or `codex` login, or `ANTHROPIC_API_KEY` / `OPENAI_API_KEY`.
- A simulator build. A device `.ipa` or an `.aab` is rejected before a session starts.
- A `claude` or `codex` login, or `ANTHROPIC_API_KEY` / `OPENAI_API_KEY`.
- EAS Simulator access and an EAS login (`eas whoami`), or `--simulator local` to run on a [local simulator](#simulator) with no EAS login.
- A simulator build. A device `.ipa`, an `.aab`, an unsigned APK, or an APK without native libraries for the emulator (x86_64 on EAS Simulator, `arm64-v8a` locally on Apple silicon; a universal APK has both) is rejected before a device starts.
- Platform, bundle id, name, and version are read from the build; you pass nothing else.
- A run stops after 15 minutes, 40 screens, or 150 actions (`--max-duration-minutes`, `--max-screens`, `--max-actions`).
- Buttons that delete, pay, sign out, send, or leave the app are never tapped. `--allow-destructive "Clear cache"` allows one by label.
- Buttons that delete, pay, sign out, send, or leave the app are never tapped. `--allow-destructive "Reset filters"` allows one by its exact label.
- `--hints "…"` tells the model what to focus on or how to sign in ([details](#sign-in-and-other-app-state)).
- The hosted link lasts 7 days (`--expires-in 30d|never`). `--no-upload` keeps the report local, and `--json` prints the result for scripts.
- For a link that stays the same from run to run, pass `--alias NAME` and set `EXPLORE_PREVIEW_TOKEN` to a fixed 32-character token.
- Read a local result with `expo-explore view --dir <folder>`, `screen:list`, or `gap:list`.
- On EAS Simulator the report is hosted, and the link lasts 7 days (`--expires-in` takes 1d to 90d, or never); `--no-upload` keeps it local. On a local simulator it opens in the local Viewer (`--no-open` or `--json` skips that; `view --dir <folder>` opens it later), and `--upload` hosts it, which needs an EAS login. `--json` prints the result for scripts.
- For a link that stays the same from run to run, pass `--alias NAME` and set `EXPLORE_PREVIEW_TOKEN` to a fixed token of 32 characters from a–z and 2–7 (`LC_ALL=C tr -dc a-z2-7 </dev/urandom | head -c 32`).
- EAS Simulator sessions and hosted reports run under an EAS project named `expo-explore` in your account, created on first use (`--owner` and `--project` pick another).
- Read a local result with `npx @expo/explore view --dir <folder>`, `screen:list`, or `gap:list`.
</details>
@@ -188,7 +188,7 @@ npx @expo/explore explore:session <session-id> --env-file path/.env.eas-simulato
<details>
<summary><strong>What a run without source shows</strong></summary>
One card per screen the run reached, laid out by how many taps it took, with the tapped control on each edge. There is no before and after; that needs [CI](#set-up-ci).
One card per screen the run reached, laid out by how many taps it took, with the tapped control on each edge. There is no before and after; that needs the app's source: [`capture`](#quick-start) on your machine, or [CI](#set-up-ci).
![The Map of an exploration report for the euxy app: one card per screen the run reached, laid out by taps from the launch screen, each edge labelled with the control that was tapped](docs/screenshots/readme-exploration-light.png)
@@ -200,7 +200,7 @@ Explore runs in your repository's GitHub Actions in two ways, both from the same
### Review on every PR
Each pull request gets a comment linking its report: the screens it added, removed or changed, before and after, on each platform you capture. Paste this into Claude Code or Codex from your app's repository:
Each pull request gets a comment linking its report: the screens it added, removed or changed, before and after, on each platform or device ([target](#targets)) you capture. Paste this into Claude Code or Codex from your app's repository:
```text
Set up this app for Expo Explore CI reviews. Follow
@@ -224,9 +224,9 @@ report:
# project: acme-explore # default <repository name>-explore, created on first run
```
- Run `expo-explore setup --workflows` after enabling it to add the `deploy` job.
- The generated workflows already have the `deploy` job; it runs once `hosting: eas` is set.
- A [comparison of two commits](#compare-any-two-commits) links its hosted report from the run's summary instead of a PR comment. `capture --upload` on your machine deploys to the same project with your `eas login`.
- `expo-explore report:upload <report.zip>...` hosts reports already on your machine the same way: a review's reports (one per platform, from `capture --output` or a downloaded `explore-review` artifact) or an `explore:binary` report.
- `expo-explore report:upload <report.zip>...` hosts reports already on your machine the same way: a review's reports (one per platform or target, from `capture --output` or a downloaded `explore-review` artifact) or an `explore:binary` report.
- `EXPO_TOKEN` must be able to create the EAS project, so use a [robot user's](https://docs.expo.dev/accounts/programmatic-access/) token.
- The report is encrypted, and only a request carrying the link's token can open it. Treat the link like a shared file link; see [report access](docs/private-review.md#hosted-previews).
@@ -238,11 +238,11 @@ report:
The prompt above walks through these steps with you. To do them yourself:
1. Run `npx @expo/explore setup` in your app's directory (in a monorepo, add `--project apps/mobile`). It writes `.expo-explore/config.yaml` and the workflows.
2. Run `npx @expo/explore doctor` and fix each problem it reports, until it exits 0. `doctor --capture` captures the committed app on EAS once, to check every screen before your first PR.
2. Run `npx @expo/explore doctor` and fix each problem it reports, until it exits 0. `doctor --capture` captures the committed app once, to check every screen before your first PR (with several targets, pass `--target`).
3. Set the repository secrets `doctor` lists:
- `EXPO_TOKEN`: create it in [Expo access tokens](https://expo.dev/settings/access-tokens). For [hosted previews](#hosted-previews), use a [robot user](https://docs.expo.dev/accounts/programmatic-access/) that can create projects.
- A secret for each `{{secret:NAME}}` in your setup steps. Run `npx @expo/explore setup --workflows` after you add one.
4. Commit, merge to the default branch, and run **Expo Explore baseline** from GitHub Actions.
4. Commit and merge to the default branch. **Expo Explore baseline** runs on that push (or run it from GitHub Actions).
5. Open a PR. **Expo Explore review** runs and posts a comment with the report. To see a whole review of your branch first, run `npx @expo/explore capture` on your machine.
A failed run shows an error code and a next action in the PR comment; see [CI troubleshooting](docs/quickstart-ci.md#diagnose-a-failed-run).
@@ -264,7 +264,7 @@ Both modes use the same comparison rules and report. You can change modes later.
**To use agent mode:**
1. Create an [Anthropic](https://platform.claude.com/settings/keys) or [OpenAI](https://platform.openai.com/api-keys) API key.
2. Set `mode: agent` and an `agent.model` your account can use.
2. Set `mode: agent`, `agent.provider` (`anthropic` or `openai`), an `agent.model` your account can use, and `agent.instructions`: the path of a file in the repository that tells the agent how to use the app.
3. Store the key as the repository secret `ANTHROPIC_API_KEY` or `OPENAI_API_KEY`. A key that is not workspace-scoped also needs the `ANTHROPIC_WORKSPACE_ID` variable.
4. Export the key and run `expo-explore doctor --check-model`.
@@ -311,17 +311,19 @@ targets:
android: { platform: android, device: pixel_7, bundleId: com.example.myapp, easProfile: explore-android }
```
- A target takes the flat `app` and `device` fields it does not set; a target of another platform than `app.platform` sets its own `bundleId` and `easProfile`. `label` names it in reports (by default its device and pose).
- Ids are lowercase letters, digits and dashes. A config with `platforms` has one target per platform, named `ios` and `android`.
- A target takes the `app` fields it does not set, and `device.name` when it is of `app.platform`; a target of another platform sets its own `bundleId` and `easProfile` (an Android one's device defaults to `pixel_7`). `pose` is `closed`, `half-open` or `open`. `label` names it in reports (by default its device and pose).
- Ids are up to 32 lowercase letters, digits and dashes, start with a letter, and are not `base`, `head` or `report`. A config names at most 8 targets. A config with `platforms` has one target per platform, named `ios` and `android`.
- The workflows run one capture job per target, the PR comment has a section per target, and each report links its sibling targets' reports.
- `capture --target duo-open` reviews one target on this machine; `--platform ios` the iOS ones.
- A target can also set its own `controller` and `simulator` over `capture`'s, such as Android on agent-device beside iOS targets on Argent. `--controller` and `--simulator` override every target's for one run.
- `capture --target duo-open` reviews one target on this machine; `--platform ios` the iOS ones. On your machine a review's targets run at once (at most `--parallel N`), and targets of one platform and EAS profile share each commit's build; in CI each target is its own job. A target that fails outright costs only its own report.
- A device controller that stops answering on one screen costs that route, up to three times per capture: Explore restarts the app, goes on with the next route, and records what happened in `capture-diagnostics/controller-recoveries.jsonl`.
</details>
<details>
<summary id="controller"><strong>Device controller: agent-device or Argent</strong></summary>
Explore drives EAS Simulator with [agent-device](https://github.com/callstackincubator/agent-device) by default. To use [Argent](https://github.com/software-mansion/argent) instead, set it in the config or pass it for one run:
Explore drives the simulator with [agent-device](https://github.com/callstackincubator/agent-device) by default. To use [Argent](https://github.com/software-mansion/argent) instead, set it in the config or pass it for one run:
```yaml
capture:
@@ -335,16 +337,16 @@ npx @expo/explore explore:binary ./app.apk --controller argent
Explore pins each controller's version and starts the EAS session with it. Reports and baselines record the controller, and a review compares captures made with the same one.
Neither controller is installed with Explore. The first run that needs one downloads its pinned npm release (agent-device 1.4 MB, Argent 27 MB), checks it against the integrity hash Explore pins, and keeps it in `~/.cache/expo-explore/controllers` (`EXPLORE_CONTROLLER_CACHE` moves it). Behind a private registry, set `EXPLORE_NPM_REGISTRY` to a mirror of it.
Neither controller is installed with Explore. The first run that needs one downloads its pinned npm release (agent-device 1.6 MB, Argent 27 MB), checks it against the integrity hash Explore pins, and keeps it in `~/.cache/expo-explore/controllers` (`EXPLORE_CONTROLLER_CACHE` moves it). Behind a private registry, set `EXPLORE_NPM_REGISTRY` to a mirror of it.
Argent cannot switch the simulator's appearance yet ([software-mansion/argent#1509](https://github.com/software-mansion/argent/pull/1509)), so an Argent capture on EAS Simulator records one appearance: the one the app forces, else light. Explore warns when it skips one. On a [local simulator](#simulator) the host switches it, so both are captured.
Argent cannot switch the simulator's appearance yet ([software-mansion/argent#1509](https://github.com/software-mansion/argent/pull/1509)), so an Argent capture on EAS Simulator records one appearance: the one the app forces, else light. Explore warns when it skips one. On a [local simulator](#simulator) the host switches it, so every appearance the config asks for is captured.
</details>
<details>
<summary id="simulator"><strong>Local simulator instead of EAS Simulator</strong></summary>
To capture on your Mac instead of EAS Simulator, set it in the config or pass it for one run:
To capture on your Mac instead of EAS Simulator, pass it for one run, or set it in a config you use only on your Mac:
```yaml
capture:
@@ -355,16 +357,22 @@ capture:
npx @expo/explore capture --simulator local
```
CI runs on GitHub's Ubuntu runners and captures on EAS Simulator, so keep `simulator: local` out of the config you commit.
For each commit, Explore starts a device of its own and removes it afterwards; head and base each get one:
- **iOS:** a simulator of the configured device with Xcode's newest runtime that runs it. `DEVELOPER_DIR` picks another Xcode, such as one that adds iPhone Duo. Explore approves the app's URL scheme on it, so links open without iOS asking "Open in …?", and checks a foldable's pose by the shape of its lit display.
- **Android:** an emulator of the configured device profile (such as `pixel_7`), from the newest Google APIs system image installed for this machine (`arm64-v8a` on Apple silicon), in a private AVD folder. The APK must ship native libraries for that ABI; a universal build does.
- **Android:** an emulator of the configured device profile (such as `pixel_7`), from the newest Google APIs system image installed for this machine (`arm64-v8a` on Apple silicon), in a private AVD folder. The APK must ship native libraries for that ABI; a universal build does. It needs the SDK's emulator, platform-tools and cmdline-tools, with `ANDROID_HOME` set.
Builds still run on EAS. The app is repacked with the route observer on your machine as well, so it is never uploaded; on Android that needs the SDK's build-tools and a JDK. `capture.repack: eas` (or `--repack eas`) keeps the repack in a hosted EAS job. `doctor --simulator local` checks that Xcode and the Android SDK can start each device, and that Android can be repacked locally.
`explore:binary --simulator local` starts one the same way for the build it explores, needs no EAS login, and opens the report in the local Viewer (`--upload` hosts it).
With `--controller argent`, Explore installs Argent's tool-server on first use (with its dependencies, about 165 MB, in the controller cache) and starts one per device. The host switches the appearance, so a local Argent capture records light and dark. Reports record where they were captured, and a baseline is reused only by a review captured the same way.
For a review, builds still run on EAS. The app is repacked with the route observer on your machine as well, so it is never uploaded; on Android that needs the SDK's build-tools and a JDK. `capture.repack: eas` (or `--repack eas`) keeps the repack in a hosted EAS job. `doctor --simulator local` checks that Xcode and the Android SDK can start each device, and that Android can be repacked locally.
A review's targets run at once, the targets of a platform sharing each commit's build; `--parallel N` runs at most N at a time. A device controller that stops answering on one screen costs that route: Explore restarts the app, goes on with the next route, and records what happened in `capture-diagnostics/controller-recoveries.jsonl`. A target that fails outright costs its own report; the others still open.
With `--controller argent`, Explore installs Argent's tool-server on first use with `npm ci` from a lockfile Explore ships, so every dependency is pinned (about 160 MB, in the controller cache), and starts one per device. The host switches the appearance, so a local Argent capture records every appearance the config asks for. Reports record where they were captured, and a baseline is reused only by a review captured the same way.
A foldable folds before the app first opens (`pose` on a [target](#targets)); with agent-device on a local simulator it folds just after, since agent-device folds only inside a session. With Argent, an opened iPhone Duo is captured upright, in landscape. After a fold, Argent's reader can stay on the panel that went dark; Explore then restarts its tool-server, and if that is not enough SpringBoard as well, and opens the app again.
Each capture records what it starts on your Mac (simulators, emulators, controller daemons, tool-servers) and removes it when it ends. If a review or a local `explore:binary` run is stopped hard, the next local one stops what it left running; nothing Explore did not record is touched.
</details>
@@ -434,10 +442,10 @@ jobs:
- Pick whatever `base` the event calls for: `main` for a release branch, yesterday's tag for a nightly, the last stable tag for a release candidate.
- `scope` and `expires-in` are optional and override `exploration.scope` and `report.preview.expiresAfter` for that run. Without them a comparison captures what a PR review would (`scope: changes` captures only screens whose source changed), and the hosted link expires like a PR's.
- The run summary links each platform's report (**Open** and **zip**, like a PR comment) and, with hosting, the hosted report.
- The run summary links each target's report (**Open** and **zip**, like a PR comment) and, with hosting, the hosted report.
- `setup --workflows` installs it from 0.19.0; `npx @expo/explore@latest setup --workflows` adds it to an existing setup. The caller grants `contents: read` and `actions: read`; `secrets: inherit` passes `EXPO_TOKEN` and any app-test secrets.
- Call it once per workflow run. A called workflow runs inside its caller's run, so a second call in the same run would reuse the first one's artifact names; use separate workflows instead.
- When the baseline workflow captured `base` in the last 30 days, that capture is reused and only `head` is captured. The report artifact is kept 30 days; only a hosted link with `expires-in: never` lasts longer.
- When the baseline workflow captured `base` in the last 30 days, with the same config and Explore version, that capture is reused and only `head` is captured. The report artifact is kept 30 days; only a hosted link with `expires-in: never` lasts longer.
</details>
@@ -465,7 +473,7 @@ If the sign-in can't be typed (one-time codes, email links, Sign in with Apple o
| Needs data to show a screen | A test account that has it, and a hint saying where ("open the first project") |
| Asks for a permission | Explore denies it. Allow it by hand on a session, then run `explore:session` |
| Has a paywall | An account that already has the entitlement; purchase buttons are never tapped |
| Has a safe button that looks destructive | `--allow-destructive "Clear cache"` |
| Has a safe button that looks destructive | `--allow-destructive "Reset filters"` |
| Is large | [Raise the budgets](#options) (15 minutes, 40 screens, 150 actions by default) and use `--hints` to set priorities |
When a run stops early, `gap:list --dir <output>` says what blocked it.
@@ -491,7 +499,7 @@ The same link answers differently depending on who asks. It already carries the
- Anything that does not ask for HTML (`curl`, `fetch`, agents' URL tools) gets a Markdown index of the report, also at `/llms.txt`.
- The index links to a read-only JSON API under `/api/v1`: `/screens?status=modified,accessibility-only,added,removed`, `/screens/<id>`, `/transitions`, and `/raw` for the report file.
- Reports without source also expose `/journeys`, `/gaps`, and `/frontier`.
- Multi-platform reports take `?platform=ios|android`. Expired links answer `410`.
- A link to several targets takes `?target=<id>` and serves the first without it; `/reports.json` lists them (`?platform=ios|android` from older links still works). Expired links' report and API routes answer `410`.
```sh
curl -s "$LINK/api/v1/screens?status=modified" | jq '.items[] | {route, status, review}'
@@ -504,7 +512,7 @@ curl -s "$LINK/api/v1/screens?status=modified" | jq '.items[] | {route, status,
1. The review workflow runs on open, non-draft PRs from the repository whose author has write access. Bots such as Devin go under `ci.trustedAuthors` in `.expo-explore/config.yaml`.
2. Explore reads the diff and your router source to pick the affected screens.
3. It builds both revisions and captures each screen on EAS Simulator, reusing a baseline of the base commit when one exists.
3. It builds both revisions on EAS (reusing a build with the same native fingerprint) and captures each screen on EAS Simulator (or, on your machine, a local simulator), reusing a baseline of the base commit when one exists.
4. It compares the captures. A screen with a source change is **Modified** if its captures differ.
5. It uploads the report to GitHub Actions and posts a PR comment.
@@ -517,13 +525,16 @@ Reports and baselines are kept for 30 days and diagnostics for 7. Hosted deploym
<details>
<summary><strong>Known limitations</strong></summary>
- Android capture has had less use than iOS. The first Android device is `pixel_7`.
- Android capture has had less use than iOS. The default Android device is `pixel_7`.
- PRs from forks are not reviewed.
- Each route gets one capture per appearance (light and dark with `capture.appearance: auto`). Other states, such as modals, are planned.
- Each route gets one capture per appearance (`capture.appearance: auto`, the default, captures light and dark when the app's `userInterfaceStyle` is `automatic`). Other states, such as modals, are planned.
- The launch is captured once per appearance and compared like any screen. The comment says how long the splash stayed as it first appeared, when it first changed (a native launch screen handing over to an app-drawn one, an animation starting on it), and when the first screen settled. `capture.launch: video` adds a recording; a launch too fast to observe is reported as "no splash observed".
- On Android the splash usually shows only in the recording, which needs `ffmpeg`. The generated workflows install it.
- Hosted acceptance of agent mode with an API key is incomplete.
- With `capture.controller: argent` on EAS Simulator, each capture records one appearance (see [device controller](#controller)).
- On EAS Simulator, an opened iPhone Duo can fail to capture when Argent's reader stays on the cover panel after the fold. A [local simulator](#simulator) recovers from it.
- A local simulator runs only on your machine; the CI workflows capture on EAS Simulator.
- `explore:session` explores EAS Simulator sessions only.
</details>
+2 -2
View File
@@ -12,11 +12,11 @@ The Viewer reads the ZIP supplied by the reviewer; it has no hosted artifact dow
## Hosted previews
Hosted previews are off by default. When `report.preview.hosting: eas` is set, the review workflow's `deploy` job uploads each run's reports (one per captured platform) to an [EAS Hosting](https://docs.expo.dev/eas/hosting/introduction/) deployment in your Expo account, and the comment shows an **Explore report** link that opens it directly, in place of the **Open** drop page. The GitHub artifact and **zip** stay as they are.
Hosted previews are off by default. When `report.preview.hosting: eas` is set, the review workflow's `deploy` job uploads each run's reports (one per captured platform or target) to an [EAS Hosting](https://docs.expo.dev/eas/hosting/introduction/) deployment in your Expo account, and the comment shows an **Explore report** link that opens it directly, in place of the **Open** drop page. The GitHub artifact and **zip** stay as they are.
**What a reviewer needs.** Only the link. It looks like `https://acme-explore--k7f3z0d1q8.expo.app/p/f6r2…` and carries a 160-bit token generated for that run. It opens the Viewer and the report for one PR revision; it does not show the repository, other runs, or anything else in the Expo account.
**Several platforms.** One deployment holds every platform's report. The link opens the primary platform's report (`?platform=ios` or `?platform=android` selects one); the Viewer's platform menu lists the others from `/p/<token>/reports.json`, which names each platform, its device, and its change count and nothing else. Every report and the listing sit behind the same token and expiry.
**Several platforms or targets.** One deployment holds every target's report (a config without `targets` has one per platform). The link opens the first one (`?target=<id>` selects another, and `?platform=ios` or `?platform=android` from older links still works); the Viewer's menu lists the others from `/p/<token>/reports.json`, which names each target, its platform, label, device, and change count and nothing else. Every report and the listing sit behind the same token and expiry.
**What is protected, and how.**
+1 -1
View File
@@ -144,7 +144,7 @@ targets:
android: { platform: android, device: pixel_7, bundleId: com.example.myapp, easProfile: explore-android }
```
A target's id (lowercase letters, digits and dashes; not `base`, `head` or `report`) takes the place of the platform everywhere a capture is named: `--target duo-open`, `explore-capture-<run>-<attempt>-duo-open`, `explore-baseline-<sha>-duo-open`. `expo-explore config:targets` prints the list the workflows use as their matrix. Fields a target leaves out come from `app` and `device`; a target of another platform than `app.platform` sets its own `bundleId` and `easProfile`. A config with `platforms` has one target per platform, named `ios` and `android`, so its names do not change. `--platform` still selects a target when only one has that platform.
A target's id (up to 32 lowercase letters, digits and dashes, starting with a letter; not `base`, `head`, `report` or `install-…`; at most 8 per config) takes the place of the platform everywhere a capture is named: `--target duo-open`, `explore-capture-<run>-<attempt>-duo-open`, `explore-baseline-<sha>-duo-open`. `expo-explore config:targets` prints the list the workflows use as their matrix. Fields a target leaves out come from `app` and `device`; a target of another platform than `app.platform` sets its own `bundleId` and `easProfile`. A config with `platforms` has one target per platform, named `ios` and `android`, so its names do not change. `--platform` still selects a target when only one has that platform.
Use the actual bundle identifier, scheme, profile, and available simulator device name. `scheme` is optional when static app configuration supplies it. `navigation.entry` resolves ambiguous entry modules; it does not list screens.
+7 -7
View File
@@ -1,6 +1,6 @@
---
name: explore-binary
description: Explore a prebuilt mobile app archive (Android .apk, iOS simulator .app, .tar.gz, or simulator .ipa) with Expo Explore. Runs `expo-explore explore:binary`, which installs the archive in a fresh EAS Simulator session, drives it with the local claude or codex login, writes an exploration map and report, hosts the report on EAS Hosting, and prints the link. Use when asked to "run @expo/explore for some.app/.apk", to map or smoke-test an app you only have as a build, or to get screenshots of every screen of a binary. `expo-explore explore:session` does the exploration half alone against an EAS Simulator session you already started (app already installed; result opens in the local Viewer). Also explores an Expo project from its source with no new build: its development build (or Expo Go) on an EAS Simulator, loading JavaScript from a tunneled local dev server. Not for device .ipa files.
description: Explore a prebuilt mobile app archive (Android .apk, iOS simulator .app, .tar.gz, or simulator .ipa) with Expo Explore. Runs `expo-explore explore:binary`, which installs the archive in a fresh EAS Simulator session (or, with `--simulator local`, on a simulator or emulator it starts on this Mac, with no EAS login), drives it with the local claude or codex login, writes an exploration map and report, and hosts the report on EAS Hosting or opens it in the local Viewer. Use when asked to "run @expo/explore for some.app/.apk", to map or smoke-test an app you only have as a build, or to get screenshots of every screen of a binary. `expo-explore explore:session` does the exploration half alone against an EAS Simulator session you already started (app already installed; result opens in the local Viewer). Also explores an Expo project from its source with no new build: its development build (or Expo Go) on an EAS Simulator, loading JavaScript from a tunneled local dev server. Not for device .ipa files.
---
# Explore a prebuilt app
@@ -11,25 +11,25 @@ One command, run from any directory:
npx @expo/explore explore:binary ./path/to/app.apk --json
```
Accepted inputs: Android `.apk`; iOS simulator builds as a `.app` directory, a `.tar.gz`/`.tgz` of one, or an `.ipa` that contains a simulator build. An iOS device `.ipa` and an Android `.aab` are rejected before any session starts (`ARCHIVE_NOT_SIMULATOR`, `ARCHIVE_AAB_UNSUPPORTED`); ask for a simulator build or a universal/x86_64 APK instead.
Accepted inputs: Android `.apk`; iOS simulator builds as a `.app` directory, a `.tar.gz`/`.tgz` of one, or an `.ipa` that contains a simulator build. An iOS device `.ipa` and an Android `.aab` are rejected before any device starts (`ARCHIVE_NOT_SIMULATOR`, `ARCHIVE_AAB_UNSUPPORTED`); ask for a simulator build or a universal APK instead.
Platform, bundle id, app name and version are read from the archive; do not pass them unless Explore asks.
## Before running
- EAS login: `eas whoami` succeeds, or `EXPO_TOKEN` is set. The session and the hosted report live under that account (`--owner` picks one when the login has several, `--project` renames the default `expo-explore` project).
- Where it runs. By default on EAS Simulator, which needs an EAS login: `eas whoami` succeeds, or `EXPO_TOKEN` is set. The session and the hosted report live under that account (`--owner` picks one when the login has several, `--project` renames the default `expo-explore` project). Without EAS Simulator access, or without an Expo account, pass `--simulator local`: Explore creates a simulator (iOS, needs Xcode) or an emulator (Android, needs the Android SDK with platform-tools, cmdline-tools, the emulator and a system image for this Mac, and `ANDROID_HOME`) and removes it afterwards. An APK must then ship native libraries for this Mac's ABI (`arm64-v8a` on Apple silicon); a universal one does.
- A model: Explore tries `ANTHROPIC_API_KEY` / `OPENAI_API_KEY` when set, then the local `claude` login (Claude Code), then the local `codex` login, and uses the first that passes a preflight request. `--provider` or `--model` limits it to that one choice. If none passes, the run stops before a simulator is started.
- Node 24.12 or newer.
## What happens
1. Archive inspection and compatibility checks (size, simulator build, emulator ABI, sha256).
2. `eas simulator:start` of an empty agent-device session (an `argent` one with `--controller argent`), verified by exact session id before use.
3. `agent-device install` of the archive straight into the session (Argent: `reinstall-app`), then `open <bundle id>`.
2. `eas simulator:start` of an empty agent-device session (an `argent` one with `--controller argent`), verified by exact session id before use. With `--simulator local`: a new simulator or emulator on this Mac instead, with the app's URL schemes approved so its links open without a prompt.
3. `agent-device install` of the archive straight into the session (Argent: `reinstall-app`; locally `simctl install` or `adb install`), then `open <bundle id>`.
4. The model explores through a small tool set (observe, act, capture, finish); app text is treated as untrusted and destructive controls (delete, reset, sign out, purchase, send, share) are blocked outside the model.
5. The map and screenshots are written under `--output` (default `./explore-<archive>-<timestamp>`), plus `report/` for the Viewer.
6. The report is sealed and deployed to EAS Hosting (`--no-upload` skips this; `--expires-in 7d|30d|never`).
7. The exact session is stopped; a failure to stop reports `SESSION_CLEANUP_FAILED` with the id to stop by hand.
6. The report is sealed and deployed to EAS Hosting (`--no-upload` skips this; `--expires-in 7d|30d|never`). With `--simulator local` it opens in the local Viewer instead, which keeps running until stopped; `--json` and `--no-open` skip it, so with `--json` open it afterwards with `view --dir <output>` (in the background, since it serves until stopped). `--upload` hosts it, and that needs an EAS login.
7. The exact session is stopped, or the local device deleted; a failure to stop reports `SESSION_CLEANUP_FAILED` with the command to stop it by hand. A local device a hard-stopped run left behind is removed by the next local run.
Budgets default to 15 minutes, 40 screens, 150 actions (`--max-duration-minutes`, `--max-screens`, `--max-actions`). Steer the model with `--hints "focus on the onboarding and settings"`, skip onboarding with `--launch-url myapp://home`, and allow a specific risky control with `--allow-destructive "Clear cache"`.
+6 -5
View File
@@ -5,7 +5,7 @@ description: Set up Expo Explore CI reviews in deterministic or agent mode. Star
# Set up Expo Explore
Make the app ready for CI capture in the user's chosen mode, using its real configuration and source. This is the shared **local authoring skill** for both modes. Android capture is not available yet.
Make the app ready for CI capture in the user's chosen mode, using its real configuration and source. This is the shared **local authoring skill** for both modes.
Explore's tools are app-agnostic: `expo-explore setup` and `expo-explore doctor` know routes, parameters, links and source redirects, but not what the app means by them. **Your job is the app-specific part**: read this app's source to learn which flows (sign-in, onboarding, a subscription, a feature flag, a permission, a record that must exist, …) each screen needs, and write that into this app's `config.yaml` or instructions. Look for what the app actually has; do not assume the flows of another app, and do not impose an account or persona schema.
@@ -164,11 +164,12 @@ npx @expo/explore capture --base v1.3.0 --head v1.4.0 --scope all
- It runs both commits' code as the user (dependency installs, the app config, bundling). Before reviewing a branch the user did not write, say so and ask; credential-free installs are not a sandbox.
- Logins: `EXPO_TOKEN` and the provider's API key when they are set; otherwise the user's `eas login` and their local `claude` or `codex` login for the configured provider. If `eas whoami` fails, tell them to run `eas login`.
- Without `--head`, the head is HEAD plus the working tree's changes, untracked non-ignored files included, which is what `eas build` uploads. It lists them; check that list for secrets the user would not want built. `--committed` reviews HEAD alone, and an `eas.json` with `cli.requireCommit: true` leaves the changes out unless `--uncommitted` is passed.
- Every configured platform is reviewed unless `--platform` picks one. It opens the report in a local Viewer; `--upload` hosts it instead and prints the link, `--no-open` or `--json` only prints the report paths.
- Every configured platform or target is reviewed, at once, unless `--target` or `--platform` picks some (`--parallel N` caps how many run together). It opens the report in a local Viewer; `--upload` hosts it instead and prints the link, `--no-open` or `--json` only prints the report paths.
- Without EAS Simulator access, add `--simulator local` (or `capture.simulator: local`): each commit is captured on a simulator Explore creates on this Mac (Xcode) or an emulator (Android SDK), and the app is repacked there too. The native build still comes from EAS. `npx @expo/explore doctor --simulator local` checks that this machine can start each device.
- Every run is kept on this machine for 30 days: `npx @expo/explore reports list` (or `--json`) shows them, `reports open <id|latest>` reopens one in the local Viewer, and `reports upload <id>` hosts it without capturing again. Run bare in a terminal, `reports` is an arrow-key picker for the user.
- To host a report after the fact, run `npx @expo/explore report:upload <report.zip>...` with the paths it printed (or the `--output` directory's `<platform>/report.zip`), one per platform. It uses the same project as `--upload` and does not capture again.
- Captures are cached in `~/.cache/expo-explore`, so a second review of the same base does not capture it again (`--no-cache` skips the cache). Ctrl+C stops the EAS Simulator sessions the review started.
- It builds on EAS and runs simulator sessions, like a CI review: expect several minutes and EAS usage per run.
- To host a report after the fact, run `npx @expo/explore report:upload <report.zip>...` with the paths it printed (or the `--output` directory's `<target>/report.zip`), one per platform or target. It uses the same project as `--upload` and does not capture again.
- Captures are cached in `~/.cache/expo-explore`, so a second review of the same base does not capture it again (`--no-cache` skips the cache). Ctrl+C stops the simulator sessions and local devices the review started; after a harder stop, the next local review stops what was left running.
- It builds on EAS and, unless it captures locally, runs EAS Simulator sessions, like a CI review: expect several minutes and EAS usage per run.
## Compare any two commits
+3 -3
View File
@@ -1,6 +1,6 @@
---
name: readme-refresh
description: Refresh Expo Explore's README.md with euxy (github.com/brentvatne/euxy) as the reference app. Regenerate every screenshot in docs/screenshots from a current euxy review report and PR comment, then re-read the README as a first-time user: the top level only says what Explore does, links the live euxy example, gives numbered quick-start steps for reviewing a change to your own app, a prompt to copy for exploring any build, and one for setting up CI; everything else lives in collapsibles. Use after a Viewer or PR-comment change that the README shows, a new command or setting, or a release that made a README claim stale. Not for the docs/ guides or the shipped skills.
description: Refresh Expo Explore's README.md with euxy (github.com/brentvatne/euxy) as the reference app. Regenerate every screenshot in docs/screenshots from a current euxy review report and PR comment, then re-read the README as a first-time user: the top level only says what Explore does, links the live euxy example, gives one quick-start prompt to copy that reviews a change to your own app or explores a build you point it to, and one for setting up CI; everything else lives in collapsibles. Use after a Viewer or PR-comment change that the README shows, a new command or setting, or a release that made a README claim stale. Not for the docs/ guides or the shipped skills.
---
# Refresh the README
@@ -60,8 +60,8 @@ Look at every PNG you produced before committing, light and dark: the selected s
Read the README top to bottom once without opening a collapsible. At that level it may only answer, in this order:
1. What Explore is (one short paragraph), the hero image, and the link to the live euxy example (`explore-example--euxy.expo.app`, redeployed on every merge to main by `.github/workflows/example-report.yml`). Under it, one table row of three pictures with a one-line caption each: Compare, Screens, and the PR comment (which says it needs CI), then the agent prompt collapsed as **Ask an agent about a report**. The hero is a zoomed Map framed as a window on a stage (`stagedWindow` in the script); the three below are plain.
2. **Quick start**: the smallest set of steps to see a review of the reader's own Expo app on their machine, numbered and with nothing optional in them: make the app buildable on EAS (login, init, bundle identifier, a simulator profile named `explore`) and commit (the before build is the last commit), then edit a screen near its top and run `capture`, which reviews uncommitted changes as `eas build` would upload them and uses setup's defaults when there is no config. After the steps, a prompt that has Claude Code or Codex do the same through the explore-setup skill's "Review a change on this machine" section. Then the `reports` picker image (`readme-reports-*.png`) with one sentence on saved runs, at the top level, and one collapsible for other commits (`--committed`, `--base`/`--head`) and what a review does on the machine. Run the steps on a fresh `create-expo-app` before changing them.
3. **Map every screen of any app**: one sentence and a prompt to paste into Claude Code or Codex that gets a simulator build of any app (Expo or not, or just a build) and runs `explore:binary` on it. Each section keeps its own "Run it yourself"; never put one section between another's prompt and its collapsibles.
2. **Quick start**: one path. A single prompt to paste into Claude Code or Codex that reviews a change to the reader's own Expo app through the explore-setup skill's "Review a change on this machine" section, or, when they point it to a build (or the app is not an Expo project), explores a simulator build through the explore-binary skill. Then one sentence on what each case needs (a review needs an Expo account, is built on EAS, and is captured on EAS Simulator or a local simulator; exploring a build needs a Claude Code or Codex login and EAS Simulator or a local simulator, and no Expo account locally). Then one **Run it yourself** collapsible with both by hand: the numbered steps for a review, with nothing optional in them (make the app buildable on EAS: login, init, bundle identifier, a simulator profile named `explore`, and commit, since the before build is the last commit; then edit a screen near its top and run `capture`, or `capture --simulator local`), and the `explore:binary` commands, including `--simulator local`. Then the `reports` picker image (`readme-reports-*.png`) with one sentence on saved runs, at the top level, and one collapsible for other commits (`--committed`, `--base`/`--head`) and what a review does on the machine. Run the steps on a fresh `create-expo-app` before changing them.
3. **Map every screen of any app**: one sentence on what exploring a build shows (pointing back to the quick start, which runs it), then its collapsibles: its options, exploring a session already started, and what a run without source shows.
4. **Set up CI**: one sentence, then two subsections, each a sentence on what it gives and a prompt to paste: **Review on every PR** (the setup prompt, then a few grouped collapsibles: Hosted previews first, then Manual setup (the setup prompt's steps done by hand, in order: setup, doctor, secrets, baseline, first PR), Capture modes: deterministic or agent, iOS, Android, or both, Keep your setup current) and **Compare any two commits** (a prompt that follows the explore-setup skill's "Compare any two commits" section, the ways to snapshot a release as bullets, then the comparison workflow collapsed).
Lead with prompts, not explanations: a reader should be able to copy something and see Explore work. Everything else, the manual commands, requirements, options, the manual setup, capture modes, hosted previews, iOS and Android, agent mode, repository settings, how agents read a report, how it works, limitations, contributing, goes into a `<details>` block under the section it belongs to, with a `<summary>` that says what the reader will find. Keep bullets to one or two sentences. Group related collapsibles rather than adding one per topic, so a section shows a handful, not nine. Requirements sit collapsed right under the example report link. Keep the collapsibles' `id` attributes that other docs link to (`#hosted-previews`, `#platforms`, `#compare`, `#agents`); the Terms and How agents read a report collapsibles live under Additional information, and keep the Inspiration section (screenmap and Revyl's Atlas).
+6 -3
View File
@@ -164,10 +164,13 @@ export async function readArchiveIcon(archive: InspectedArchive): Promise<Buffer
}
}
/** Explore's EAS Android emulator is x86_64 with no 32-bit support; an APK that ships only ARM or x86 libraries fails to install there. */
/**
* Explore's EAS Android emulator is x86_64 with no 32-bit support, and a local one runs this Mac's ABI; an APK that
* ships native libraries only for other ABIs fails to install there.
*/
export function assertEmulatorCompatible(archive: InspectedArchive, emulatorAbis: readonly string[] = ["x86_64"]): void {
if (archive.platform === "ios") {
if (!archive.simulator) throw archiveError("ARCHIVE_NOT_SIMULATOR", `${basename(archive.path)} is an iOS device build; it cannot run on the EAS Simulator.`,
if (!archive.simulator) throw archiveError("ARCHIVE_NOT_SIMULATOR", `${basename(archive.path)} is an iOS device build; it cannot run on a simulator.`,
"Build for the simulator (an EAS profile with ios.simulator: true) and pass that .app, .tar.gz, or simulator .ipa.");
return;
}
@@ -177,7 +180,7 @@ export function assertEmulatorCompatible(archive: InspectedArchive, emulatorAbis
}
if (archive.abis.length && !archive.abis.some(abi => emulatorAbis.includes(abi))) {
throw archiveError("ARCHIVE_ABI_UNSUPPORTED", `The APK ships native libraries only for ${archive.abis.join(", ")}; the emulator needs ${emulatorAbis.join(" or ")}.`,
"Build a universal APK or one that includes x86_64 native libraries.", { abis: archive.abis, emulatorAbis: [...emulatorAbis] });
`Build a universal APK or one that includes ${emulatorAbis[0]} native libraries.`, { abis: archive.abis, emulatorAbis: [...emulatorAbis] });
}
}
+168 -63
View File
@@ -13,7 +13,10 @@ import { executeCaptureSetup } from "../capture/setup-actions.js";
import { SIMULATOR_DOTENV_FILE_NAME, SIMULATOR_DOTENV_HEADER, assertSimulatorSessionIdentity, assertSimulatorStopped, readSimulatorDotenvHandoff,
assertAttachedSimulatorSession, connectionController, controllerVersion, resolveAttachedSimulatorSession, simulatorOsVersion, simulatorStartFailure, type ActiveSimulatorRuntime, type SimulatorConnection } from "../capture/simulator.js";
import { argentControllerInvocation } from "../capture/controller-invocation.js";
import { agentDeviceLauncher, ensureController } from "../shared/controller-cache.js";
import { agentDeviceLauncher, ensureArgentToolServer, ensureController } from "../shared/controller-cache.js";
import { agentDeviceClaims, approveLinkScheme, createLocalSimulator, deleteLocalEmulator, deleteLocalSimulator, hostEmulatorAbi, installOnLocalEmulator, installOnLocalSimulator, startLocalArgentServer,
startLocalEmulator, stopLocalAgentDevice, stopLocalArgentServer, waitForLocalEmulator, waitForLocalSimulator, type LocalArgentServer } from "../capture/local-simulator.js";
import { openLocalCaptureRecord, sweepLocalCaptures } from "../capture/local-registry.js";
import { DEFAULT_ANDROID_DEVICE } from "../config/explore-config.js";
import { createExplorationInputs } from "../config/exploration-inputs.js";
import { ExploreStore, getCoverage } from "../map/store.js";
@@ -50,6 +53,8 @@ export type BinaryExploreOptions = {
project?: string;
/** The session's device controller; agent-device unless Argent is chosen. explore:session takes it from the session. */
controller?: "agent-device" | "argent";
/** Where the app runs: an EAS Simulator session (the default), or a simulator or emulator this run starts on this machine, with no EAS login. */
simulator?: "eas" | "local";
environment?: NodeJS.ProcessEnv;
runner?: typeof defaultRunner;
createSession?: (config: CiAgentSessionConfig, options: { env: NodeJS.ProcessEnv; auth: ModelAuth }) => CiAgentSession;
@@ -71,8 +76,8 @@ export type BinaryExploreResult = {
/** The command and environment that reach a session's controller: agent-device's CLI with its daemon, or the Argent bridge. */
type ControllerInvocation = { command: string; environment: NodeJS.ProcessEnv };
async function argentInvocation(connection: SimulatorConnection, work: string, platform: Platform, environment: NodeJS.ProcessEnv, bundleId?: string): Promise<ControllerInvocation> {
return argentControllerInvocation({ directory: work, connection, platform, environment: baseEnvironment(environment), sourceEnvironment: environment, ...(bundleId ? { bundleId } : {}) });
async function argentInvocation(connection: SimulatorConnection, work: string, platform: Platform, environment: NodeJS.ProcessEnv, bundleId?: string, device?: string): Promise<ControllerInvocation> {
return argentControllerInvocation({ directory: work, connection, platform, environment: baseEnvironment(environment), sourceEnvironment: environment, ...(bundleId ? { bundleId } : {}), ...(device ? { device } : {}) });
}
/** The daemon refuses a second session on a busy device; the caller's own `agent-device` session (from install/open) is reused instead. */
@@ -267,11 +272,16 @@ export type ConnectedSessionExploration = {
urlSchemes?: readonly string[];
/** The app's icon from the archive for the Viewer header. */
appIcon?: Buffer;
/**
* A simulator or emulator on this machine instead of an EAS session: the controller drives it by its UDID or adb
* serial (agent-device with a daemon of its own in `stateDirectory`, or a local Argent tool-server pinned to it).
*/
local?: { deviceId: string; osVersion: string; stateDirectory: string };
};
/** Drives an agent-device daemon we already hold a connection to: optional install, open, then the model-led exploration. */
export async function exploreConnectedSession(ctx: ConnectedSessionExploration): Promise<{ runtime: ActiveSimulatorRuntime; exploration: BlackBoxExplorerResult }> {
const { connection, platform, device, bundleId, app, install, work, runner, environment, secrets, inputs, say, now, limits, store, map, model, sessionJson, launchUrl, hints, allowDestructive } = ctx;
const { connection, platform, device, bundleId, app, install, work, runner, environment, secrets, inputs, say, now, limits, store, map, model, sessionJson, launchUrl, hints, allowDestructive, local } = ctx;
const controllerSession = ctx.controllerSession ?? CONTROLLER_SESSION;
const controllerKind = connectionController(connection);
const controllerName = controllerKind === "argent" ? "Argent" : "agent-device";
@@ -279,15 +289,18 @@ export async function exploreConnectedSession(ctx: ConnectedSessionExploration):
let agentDevice: string;
let controllerEnvironment: NodeJS.ProcessEnv;
if (controllerKind === "argent") {
({ command: agentDevice, environment: controllerEnvironment } = await argentInvocation(connection, work, platform, environment, bundleId));
({ command: agentDevice, environment: controllerEnvironment } = await argentInvocation(connection, work, platform, environment, bundleId, local?.deviceId));
} else {
const controllerConfig = join(work, "agent-device.json");
await writeFile(controllerConfig, "{}\n", { flag: "wx", mode: 0o400 });
controllerEnvironment = { ...baseEnvironment(environment), AGENT_DEVICE_CONFIG: controllerConfig, AGENT_DEVICE_DAEMON_BASE_URL: connection.daemonBaseUrl,
AGENT_DEVICE_DAEMON_AUTH_TOKEN: connection.daemonAuthToken, AGENT_DEVICE_SESSION: controllerSession, AGENT_DEVICE_SESSION_LOCK: "reject", AGENT_DEVICE_PLATFORM: platform };
// A local device has no EAS daemon: agent-device runs its own, in a state directory of this run's.
controllerEnvironment = { ...baseEnvironment(environment), AGENT_DEVICE_CONFIG: controllerConfig,
...(local ? { AGENT_DEVICE_STATE_DIR: local.stateDirectory } : { AGENT_DEVICE_DAEMON_BASE_URL: connection.daemonBaseUrl, AGENT_DEVICE_DAEMON_AUTH_TOKEN: connection.daemonAuthToken }),
AGENT_DEVICE_SESSION: controllerSession, AGENT_DEVICE_SESSION_LOCK: "reject", AGENT_DEVICE_PLATFORM: platform };
agentDevice = agentDeviceLauncher(environment);
}
const deviceArgs = platform === "ios" ? ["--device", device] : [];
// A local device is named for Explore, so it is picked by its UDID or serial rather than its model name.
const deviceArgs = local ? [platform === "ios" ? "--udid" : "--serial", local.deviceId] : platform === "ios" ? ["--device", device] : [];
const controller = (args: string[], timeoutMs = 90_000): CommandResult => {
// Session flags go before a `--` separator so model-chosen text (`type -h`, a URL, a label) is never read as a flag.
const separator = args.indexOf("--");
@@ -304,16 +317,17 @@ export async function exploreConnectedSession(ctx: ConnectedSessionExploration):
try { controller(["install", install.path, "--json"], 15 * 60_000); }
catch (error) { throw binaryError("APP_INSTALL_FAILED", `${controllerName} could not install the app: ${failureText(error)}`, "Check that the archive is a simulator build for this platform and under the size limit, then rerun.", { sessionId: connection.sessionId }); }
}
const opened = controller(["open", bundleId, "--level", "full", "--json"], 2 * 60_000);
// On a local device, the first open also builds agent-device's iOS runner.
const opened = controller(["open", bundleId, "--level", "full", "--json"], local ? 15 * 60_000 : 2 * 60_000);
const openedDevice = openedSimulatorDevice(opened.stdout, bundleId, "initial app open", platform, controllerSession);
const data = controllerData(opened.stdout, "initial app open");
const deviceData = data.device && typeof data.device === "object" ? data.device as Record<string, unknown> : {};
const observed = (key: string): string => { const value = deviceData[key] ?? data[key]; return typeof value === "string" && value.trim() ? value : "unknown"; };
const runtime = { provider: "eas-simulator", controller: controllerKind, controllerPackageVersion: controllerVersion(controllerKind), easCliVersion: "unknown", platform, controllerDeviceId: openedDevice.deviceId,
deviceModel: openedDevice.deviceModel, osVersion: observed("osVersion"), locale: observed("locale"), orientation: observed("orientation") === "landscape" ? "landscape" : "portrait",
const runtime = { provider: local ? "local-simulator" : "eas-simulator", controller: controllerKind, controllerPackageVersion: controllerVersion(controllerKind), easCliVersion: "unknown", platform, controllerDeviceId: openedDevice.deviceId,
deviceModel: local ? device : openedDevice.deviceModel, osVersion: local ? local.osVersion : observed("osVersion"), locale: observed("locale"), orientation: observed("orientation") === "landscape" ? "landscape" : "portrait",
colorScheme: observed("colorScheme") === "dark" ? "dark" : "light" } as ActiveSimulatorRuntime;
// Argent's `open` reports the OS version itself (list-devices); agent-device's comes from the session's preview.
if (runtime.osVersion === "unknown" || controllerKind !== "argent") try { runtime.osVersion = await simulatorOsVersion(sessionJson(), runtime.controllerDeviceId, fetch, platform); } catch { /* descriptive only */ }
if (!local && (runtime.osVersion === "unknown" || controllerKind !== "argent")) try { runtime.osVersion = await simulatorOsVersion(sessionJson(), runtime.controllerDeviceId, fetch, platform); } catch { /* descriptive only */ }
map.run.deviceId = runtime.controllerDeviceId;
map.run.environment = { deviceModel: runtime.deviceModel, osVersion: runtime.osVersion, locale: runtime.locale, orientation: runtime.orientation, colorScheme: runtime.colorScheme };
map.run.status = "exploring";
@@ -572,6 +586,75 @@ export async function runSessionExploration(options: SessionExploreOptions): Pro
}
}
type LocalDevice = NonNullable<ConnectedSessionExploration["local"]>;
/**
* Starts a simulator or emulator on this machine for one run, installs the app, and hands the explorer a connection
* to it; everything is stopped afterwards. They are the devices `capture --simulator local` starts, recorded the same
* way, so the next local run stops what a run stopped hard left behind.
*/
async function withLocalDevice(options: { archive: InspectedArchive; installPath: string; platform: Platform; device: string; bundleId: string;
controller: "agent-device" | "argent"; tag: string; work: string; runner: typeof defaultRunner; environment: NodeJS.ProcessEnv; secrets: readonly string[]; say: (message: string) => void; now?: () => number },
explore: (connection: SimulatorConnection, local: LocalDevice) => Promise<void>): Promise<void> {
const { archive, installPath, platform, device, bundleId, controller, tag, work, runner, say } = options;
const environment = baseEnvironment(options.environment);
const agentDevice = agentDeviceLauncher(options.environment);
const stateDirectory = join(work, "agent-device-state");
const record = await openLocalCaptureRecord(options.environment, work);
let udid: string | undefined, serial: string | undefined, argent: LocalArgentServer | undefined;
let primaryError: unknown;
try {
let osVersion: string;
if (platform === "ios") {
say(`starting a local ${device} simulator`);
const created = await createLocalSimulator({ runner, device, name: `Explore ${device} (binary ${tag})`, cwd: work, environment,
created: async id => { udid = id; await record.update({ udid: id }); } });
osVersion = created.osVersion;
waitForLocalSimulator({ runner, udid: created.udid, cwd: work, environment });
// Links in the app's own schemes open without iOS asking first.
for (const scheme of archive.urlSchemes) approveLinkScheme({ runner, udid: created.udid, scheme, bundleId, cwd: work, environment });
say("installing the app");
installOnLocalSimulator({ runner, udid: created.udid, app: installPath, cwd: work, environment });
} else {
say(`starting a local ${device} emulator`);
const name = `explore-binary-${tag}`;
const started = await startLocalEmulator({ runner, device, name, directory: work, cwd: work, environment,
...(controller === "agent-device" ? { claimed: (candidate: string) => agentDeviceClaims({ runner, agentDevice, serial: candidate, cwd: work, environment }) } : {}),
started: async id => { serial = id; await record.update({ serial: id, avd: name }); } });
osVersion = waitForLocalEmulator({ runner, serial: started.serial, device, apiLevel: started.apiLevel, directory: work, cwd: work, environment, ...(options.now ? { now: options.now } : {}) });
say("installing the APK");
installOnLocalEmulator({ runner, serial: started.serial, apk: installPath, cwd: work, environment });
}
const deviceId = (udid ?? serial)!;
let connection: SimulatorConnection = { sessionId: `local-${deviceId}`, daemonBaseUrl: "", daemonAuthToken: "" };
if (controller === "argent") {
// Argent drives the device through a tool-server of this run's own.
const cli = await ensureArgentToolServer({ environment: options.environment, progress: say });
argent = await startLocalArgentServer({ runner, cli, directory: work, cwd: work, environment, ...(options.now ? { now: options.now } : {}) });
await record.update({ argent: { pid: argent.pid, cli: argent.cli, home: argent.home } });
connection = { ...connection, daemonBaseUrl: argent.url, daemonAuthToken: argent.token, controller: "argent" };
}
await explore(connection, { deviceId, osVersion, stateDirectory });
} catch (error) { primaryError = error; throw error; }
finally {
try {
if (argent) stopLocalArgentServer(argent);
// Its own agent-device daemon lets go of the device before it goes.
if (controller === "agent-device") runner({ command: agentDevice, args: ["close", "--session", CONTROLLER_SESSION, "--json"], cwd: work,
environment: { ...environment, AGENT_DEVICE_STATE_DIR: stateDirectory }, timeoutMs: 60_000, allowFailure: true });
await stopLocalAgentDevice(stateDirectory);
if (serial) { say(`stopping the local emulator ${serial}`); await deleteLocalEmulator({ runner, serial, cwd: work, environment }); }
if (udid) { say(`deleting the local simulator ${udid}`); deleteLocalSimulator({ runner, udid, cwd: work, environment }); }
// Everything is stopped; a cleanup that threw leaves the record for the next run's sweep.
await record.close();
} catch (error) {
throw new CiError({ code: "SESSION_CLEANUP_FAILED", stage: "cleanup", message: `The local ${serial ? "emulator" : "simulator"} Explore started could not be stopped.`,
nextAction: serial ? `Stop it with: adb -s ${serial} emu kill` : udid ? `Delete it with: xcrun simctl delete ${udid}` : "Rerun explore:binary --simulator local; it stops what this run left.",
details: { ...(serial ? { serial } : {}), ...(udid ? { udid } : {}), ...(primaryError ? { originalFailure: serializeCiError(primaryError, { secrets: options.secrets }) } : {}) }, cause: error });
}
}
}
export async function runBinaryExploration(options: BinaryExploreOptions): Promise<BinaryExploreResult> {
const startedAt = Date.now();
const now = options.now ?? Date.now;
@@ -580,18 +663,21 @@ export async function runBinaryExploration(options: BinaryExploreOptions): Promi
const runner = options.runner ?? defaultRunner;
const limits = validatedLimits(options);
const inputs = hintInputs(options.hints, environment);
const local = options.simulator === "local";
// A local emulator runs this Mac's ABI (arm64-v8a on Apple silicon); EAS's runs x86_64.
const emulatorAbi = local ? hostEmulatorAbi() : "x86_64";
say("inspecting the archive");
const archive = await inspectArchive(options.archive);
assertEmulatorCompatible(archive);
assertEmulatorCompatible(archive, [emulatorAbi]);
if (options.platform && options.platform !== archive.platform) throw binaryError("ARCHIVE_PLATFORM_MISMATCH", `The archive is an ${archive.platform} build but --platform ${options.platform} was requested.`, "Drop --platform or pass a matching archive.");
if (!environment.EXPO_TOKEN?.trim() && !environment.EXPO_SESSION_SECRET?.trim() && !(await easLoggedIn(runner, environment))) {
if (!local && !environment.EXPO_TOKEN?.trim() && !environment.EXPO_SESSION_SECRET?.trim() && !(await easLoggedIn(runner, environment))) {
throw binaryError("EXPO_LOGIN_REQUIRED", "No EAS login was found.", "Run `eas login` or set EXPO_TOKEN, then rerun explore:binary.");
}
const platform = archive.platform;
const bundleId = options.bundleId ?? archive.bundleId;
const device = options.device ?? (platform === "ios" ? DEFAULT_IOS_DEVICE : DEFAULT_ANDROID_DEVICE);
say(`${archive.appName} ${archive.version ?? ""} (${bundleId}, ${platform}, ${(archive.bytes / 1048576).toFixed(0)} MB, sha256 ${archive.sha256.slice(0, 12)})`);
if (archive.missingLibs?.length) say(`warning: lib/x86_64 lacks ${archive.missingLibs.slice(0, 4).join(", ")}${archive.missingLibs.length > 4 ? ` and ${archive.missingLibs.length - 4} more` : ""} that another ABI ships; the app may stop with UnsatisfiedLinkError on the emulator`);
if (archive.missingLibs?.length && emulatorAbi === "x86_64") say(`warning: lib/x86_64 lacks ${archive.missingLibs.slice(0, 4).join(", ")}${archive.missingLibs.length > 4 ? ` and ${archive.missingLibs.length - 4} more` : ""} that another ABI ships; the app may stop with UnsatisfiedLinkError on the emulator`);
const model = await preflightModelSession(options, environment, say);
@@ -601,7 +687,7 @@ export async function runBinaryExploration(options: BinaryExploreOptions): Promi
const work = await realpath(await mkdtemp(join(tmpdir(), "expo-explore-binary-")));
const easEnvironment = { ...baseEnvironment(environment), ...(environment.EXPO_TOKEN ? { EXPO_TOKEN: environment.EXPO_TOKEN } : {}), ...(environment.HOME ? { HOME: environment.HOME } : {}), CI: "1", EXPO_NO_TELEMETRY: "1" };
const easBin = packageBin("eas-cli", "eas");
await lstat(easBin).catch(() => { throw binaryError("EXPLORE_TOOLS_MISSING", "The EAS CLI that Explore pins is not installed.", "Reinstall @expo/explore with optional dependencies."); });
if (!local) await lstat(easBin).catch(() => { throw binaryError("EXPLORE_TOOLS_MISSING", "The EAS CLI that Explore pins is not installed.", "Reinstall @expo/explore with optional dependencies."); });
const eas = (args: string[], timeoutMs = 5 * 60_000): CommandResult & { status: number } => {
try { return { ...runner({ command: process.execPath, args: [easBin, ...args], cwd: work, environment: easEnvironment, timeoutMs }), status: 0 }; }
catch (error) { return { stdout: "", stderr: error instanceof Error ? error.message : String(error), status: 1 }; }
@@ -615,6 +701,11 @@ export async function runBinaryExploration(options: BinaryExploreOptions): Promi
const controller = options.controller ?? "agent-device";
// Fetched before the paid session starts, so a blocked download costs nothing.
if (!options.runner) await ensureController(controller, { environment, progress: say });
if (local) {
// What an earlier local run stopped hard left running (a simulator, an emulator, a daemon) goes first.
const swept = await sweepLocalCaptures({ environment, runner, cwd: work }).catch(() => [] as string[]);
if (swept.length) say(`stopped what an interrupted local run left running: ${swept.join(", ")}`);
}
const map: ExploreMap = { contract: CONTRACT, revision: 0, run: { id: randomUUID(), name: archive.appName, platform, controller, exploration: "black-box", bundleId, status: "draft",
createdAt: new Date(now()).toISOString(), updatedAt: new Date(now()).toISOString(),
build: { ...(archive.version ? { version: archive.version } : {}), ...(archive.buildNumber ? { number: archive.buildNumber } : {}), hash: archive.sha256, source: "local" },
@@ -622,57 +713,71 @@ export async function runBinaryExploration(options: BinaryExploreOptions): Promi
try {
let installPath = archive.path;
if (archive.kind === "ios-tgz" || archive.kind === "ios-ipa") { say("unpacking the iOS bundle"); installPath = await extractIosBundle(archive, join(work, "bundle")); }
if (!owner) {
const who = eas(["whoami"], 60_000);
owner = who.status === 0 ? whoamiAccounts(who.stdout)[0] ?? "" : "";
if (!owner) throw binaryError("EXPO_LOGIN_REQUIRED", "EAS did not report an account for this login.", "Run `eas login` (or set EXPO_TOKEN) and rerun, or pass --owner.");
}
await writeFile(join(work, "package.json"), JSON.stringify({ name: project, version: "0.0.0", private: true }), { mode: 0o600 });
await writeFile(join(work, "app.json"), JSON.stringify({ expo: { name: project, slug: project, owner, web: { output: "server" } } }), { mode: 0o600 });
const init = eas(["init", "--force", "--non-interactive", "--json"], 2 * 60_000);
const appJson = JSON.parse(await readFile(join(work, "app.json"), "utf8")) as { expo?: { extra?: { eas?: { projectId?: string } } } };
const projectId = appJson.expo?.extra?.eas?.projectId;
if (init.status !== 0 || !projectId) throw binaryError("EXPLORE_PROJECT_FAILED", `EAS could not create or link @${owner}/${project}: ${lastLine(init.stderr || init.stdout)}`,
"Give this login permission to create projects under the owner, or pass --project with an existing slug.", { owner, project });
const projectIdentity = { name: project, slug: project, owner, projectId };
say(`starting an EAS Simulator session (${device}, ${limits.maxDurationMinutes} min)`);
const sessionName = `explore-binary-${createHash("sha256").update(`${archive.sha256}:${map.run.id}`).digest("hex").slice(0, 16)}`;
const dotenvPath = join(work, SIMULATOR_DOTENV_FILE_NAME);
await writeFile(dotenvPath, SIMULATOR_DOTENV_HEADER, { flag: "wx", mode: 0o600 });
let startError: unknown;
try {
const started = eas(["simulator:start", "--name", sessionName, "--platform", platform, "--device", device, "--type", controller, "--package-version", controllerVersion(controller), "--public-link",
"--max-duration-minutes", String(limits.maxDurationMinutes + 5), "--out-config-type", "dotenv", "--non-interactive"], 20 * 60_000);
if (started.status !== 0) throw new Error(started.stderr || started.stdout || "eas simulator:start failed");
} catch (error) { startError = error; }
const handoff = await readSimulatorDotenvHandoff(dotenvPath);
const secrets = [environment.EXPO_TOKEN ?? "", ...(handoff.connection ? [handoff.connection.daemonAuthToken] : []), ...inputs.values].filter(Boolean);
const startFailure = startError ? simulatorStartFailure(startError, secrets) : undefined;
if (handoff.sessionId) {
try { assertSimulatorSessionIdentity(eas(["simulator:get", "--id", handoff.sessionId, "--json", "--non-interactive"], 2 * 60_000).stdout, handoff.sessionId, sessionName, projectIdentity, handoff.connection, platform); }
catch (error) { throw new CiError({ code: "SESSION_HANDOFF_UNVERIFIED", stage: "session", message: "The EAS handoff could not prove ownership of the exact simulator session. The private handoff was preserved.",
nextAction: "Inspect the EAS session list; confirm ownership before stopping the reported session.", details: { handoffDirectory: work, reportedSessionId: handoff.sessionId, ...(startFailure ? { startFailure } : {}) }, cause: error }); }
sessionId = handoff.sessionId;
}
if (!sessionId) throw new CiError({ code: startFailure ? "SIMULATOR_START_FAILED" : "SESSION_HANDOFF_UNVERIFIED", stage: "session",
message: startFailure ? `EAS simulator:start failed: ${startFailure.message}` : "EAS did not provide a verifiable simulator session.",
nextAction: "Inspect the EAS start failure and session list; verify a session's exact ID and ownership before cleanup.", details: { handoffDirectory: work, ...(startFailure ? { startFailure } : {}) }, cause: startError });
if (startFailure) throw new CiError({ code: "SIMULATOR_START_FAILED", stage: "session", message: `EAS simulator:start failed after the session was verified: ${startFailure.message}`,
nextAction: "Inspect the start failure; Explore stops the verified session before returning.", details: { sessionId, startFailure }, cause: startError });
if (handoff.validationError) throw handoff.validationError;
if (!handoff.connection) throw new Error("Simulator handoff is missing its controller connection");
await rm(dotenvPath, { force: true });
const appIcon = await readArchiveIcon(archive);
({ runtime, exploration } = await exploreConnectedSession({ connection: handoff.connection, platform, device, bundleId, ...(appIcon ? { appIcon } : {}), app: { name: archive.appName, ...(archive.version ? { version: archive.version } : {}) },
install: { path: installPath, kind: archive.kind }, work, runner, environment, secrets, inputs, say, now, limits, store, map, model, urlSchemes: archive.urlSchemes,
sessionJson: () => eas(["simulator:get", "--id", handoff.connection!.sessionId, "--json", "--non-interactive"], 2 * 60_000).stdout,
...(options.launchUrl ? { launchUrl: options.launchUrl } : {}), ...(options.hints ? { hints: options.hints } : {}), ...(options.allowDestructive ? { allowDestructive: options.allowDestructive } : {}) }));
const exploreOn = (connection: SimulatorConnection, sessionJson: () => string, localDevice?: LocalDevice) => exploreConnectedSession({ connection, platform, device, bundleId,
...(appIcon ? { appIcon } : {}), app: { name: archive.appName, ...(archive.version ? { version: archive.version } : {}) },
// A local device has the app installed by simctl or adb before the controller starts.
...(localDevice ? { local: localDevice } : { install: { path: installPath, kind: archive.kind } }),
work, runner, environment, secrets: [environment.EXPO_TOKEN ?? "", connection.daemonAuthToken, ...inputs.values].filter(Boolean), inputs, say, now, limits, store, map, model,
urlSchemes: archive.urlSchemes, sessionJson,
...(options.launchUrl ? { launchUrl: options.launchUrl } : {}), ...(options.hints ? { hints: options.hints } : {}), ...(options.allowDestructive ? { allowDestructive: options.allowDestructive } : {}) });
if (local) {
const tag = map.run.id.replace(/-/g, "").slice(0, 8);
await withLocalDevice({ archive, installPath, platform, device, bundleId, controller, tag, work, runner, environment, say, secrets: inputs.values, ...(options.now ? { now: options.now } : {}) },
async (connection, localDevice) => {
sessionId = connection.sessionId;
({ runtime, exploration } = await exploreOn(connection, () => "", localDevice));
});
} else {
if (!owner) {
const who = eas(["whoami"], 60_000);
owner = who.status === 0 ? whoamiAccounts(who.stdout)[0] ?? "" : "";
if (!owner) throw binaryError("EXPO_LOGIN_REQUIRED", "EAS did not report an account for this login.", "Run `eas login` (or set EXPO_TOKEN) and rerun, or pass --owner.");
}
await writeFile(join(work, "package.json"), JSON.stringify({ name: project, version: "0.0.0", private: true }), { mode: 0o600 });
await writeFile(join(work, "app.json"), JSON.stringify({ expo: { name: project, slug: project, owner, web: { output: "server" } } }), { mode: 0o600 });
const init = eas(["init", "--force", "--non-interactive", "--json"], 2 * 60_000);
const appJson = JSON.parse(await readFile(join(work, "app.json"), "utf8")) as { expo?: { extra?: { eas?: { projectId?: string } } } };
const projectId = appJson.expo?.extra?.eas?.projectId;
if (init.status !== 0 || !projectId) throw binaryError("EXPLORE_PROJECT_FAILED", `EAS could not create or link @${owner}/${project}: ${lastLine(init.stderr || init.stdout)}`,
"Give this login permission to create projects under the owner, or pass --project with an existing slug.", { owner, project });
const projectIdentity = { name: project, slug: project, owner, projectId };
say(`starting an EAS Simulator session (${device}, ${limits.maxDurationMinutes} min)`);
const sessionName = `explore-binary-${createHash("sha256").update(`${archive.sha256}:${map.run.id}`).digest("hex").slice(0, 16)}`;
const dotenvPath = join(work, SIMULATOR_DOTENV_FILE_NAME);
await writeFile(dotenvPath, SIMULATOR_DOTENV_HEADER, { flag: "wx", mode: 0o600 });
let startError: unknown;
try {
const started = eas(["simulator:start", "--name", sessionName, "--platform", platform, "--device", device, "--type", controller, "--package-version", controllerVersion(controller), "--public-link",
"--max-duration-minutes", String(limits.maxDurationMinutes + 5), "--out-config-type", "dotenv", "--non-interactive"], 20 * 60_000);
if (started.status !== 0) throw new Error(started.stderr || started.stdout || "eas simulator:start failed");
} catch (error) { startError = error; }
const handoff = await readSimulatorDotenvHandoff(dotenvPath);
const secrets = [environment.EXPO_TOKEN ?? "", ...(handoff.connection ? [handoff.connection.daemonAuthToken] : []), ...inputs.values].filter(Boolean);
const startFailure = startError ? simulatorStartFailure(startError, secrets) : undefined;
if (handoff.sessionId) {
try { assertSimulatorSessionIdentity(eas(["simulator:get", "--id", handoff.sessionId, "--json", "--non-interactive"], 2 * 60_000).stdout, handoff.sessionId, sessionName, projectIdentity, handoff.connection, platform); }
catch (error) { throw new CiError({ code: "SESSION_HANDOFF_UNVERIFIED", stage: "session", message: "The EAS handoff could not prove ownership of the exact simulator session. The private handoff was preserved.",
nextAction: "Inspect the EAS session list; confirm ownership before stopping the reported session.", details: { handoffDirectory: work, reportedSessionId: handoff.sessionId, ...(startFailure ? { startFailure } : {}) }, cause: error }); }
sessionId = handoff.sessionId;
}
if (!sessionId) throw new CiError({ code: startFailure ? "SIMULATOR_START_FAILED" : "SESSION_HANDOFF_UNVERIFIED", stage: "session",
message: startFailure ? `EAS simulator:start failed: ${startFailure.message}` : "EAS did not provide a verifiable simulator session.",
nextAction: "Inspect the EAS start failure and session list; verify a session's exact ID and ownership before cleanup.", details: { handoffDirectory: work, ...(startFailure ? { startFailure } : {}) }, cause: startError });
if (startFailure) throw new CiError({ code: "SIMULATOR_START_FAILED", stage: "session", message: `EAS simulator:start failed after the session was verified: ${startFailure.message}`,
nextAction: "Inspect the start failure; Explore stops the verified session before returning.", details: { sessionId, startFailure }, cause: startError });
if (handoff.validationError) throw handoff.validationError;
if (!handoff.connection) throw new Error("Simulator handoff is missing its controller connection");
await rm(dotenvPath, { force: true });
({ runtime, exploration } = await exploreOn(handoff.connection, () => eas(["simulator:get", "--id", handoff.connection!.sessionId, "--json", "--non-interactive"], 2 * 60_000).stdout));
}
} catch (error) {
primaryError = toCiError(error, { code: "EXPLORE_BINARY_FAILED", stage: "exploration", nextAction: "Inspect the diagnostic, correct the failing prerequisite, and rerun explore:binary." });
} finally {
if (sessionId) {
// A local device is stopped by withLocalDevice; only the work directory is left.
if (local) await rm(work, { recursive: true, force: true });
else if (sessionId) {
try {
say(`stopping simulator session ${sessionId}`);
assertSimulatorStopped(eas(["simulator:stop", "--id", sessionId, "--json", "--non-interactive"], 2 * 60_000).stdout, sessionId);
+12
View File
@@ -186,6 +186,18 @@ export async function assertLocalEmulatorCompatible(apk: string, abi: string): P
nextAction: `Build an APK that includes ${abi} (a universal build does), or capture Android on EAS Simulator with --simulator eas.` });
}
/**
* Whether agent-device still holds a claim on an emulator serial, one an interrupted run left behind: it refuses
* such a serial even when it reports the claim as stale, so startLocalEmulator skips that port.
*/
export function agentDeviceClaims(options: { runner: CommandRunner; agentDevice: string; serial: string; cwd: string; environment: NodeJS.ProcessEnv }): boolean {
const status = options.runner({ command: options.agentDevice, args: ["device", "status", "--platform", "android", "--serial", options.serial, "--json"],
cwd: options.cwd, environment: options.environment, timeoutMs: 30_000, allowFailure: true });
let data: { claims?: unknown[]; hiddenStaleClaims?: number } | undefined;
try { data = (JSON.parse(status.stdout) as { data?: typeof data }).data; } catch { return false; }
return (data?.claims?.length ?? 0) + (data?.hiddenStaleClaims ?? 0) > 0;
}
/**
* Creates an AVD of the configured device profile (such as pixel_7) and starts it without a window. It boots in the
* background while a capture repacks the app; waitForLocalEmulator waits for it afterwards.
+2 -12
View File
@@ -25,7 +25,7 @@ import { assertCaptureEasProfile, assertSimulatorAvailable, fingerprintFrom, has
import { SIMULATOR_DOTENV_FILE_NAME, SIMULATOR_DOTENV_HEADER, simulatorSessionName, simulatorStartFailure, readSimulatorDotenvHandoff, assertSimulatorSessionIdentity, assertSimulatorStopped, collectRuntime, controllerVersion, simulatorOsVersion, type ActiveSimulatorRuntime, type SimulatorConnection } from "./simulator.js";
import { argentControllerInvocation } from "./controller-invocation.js";
import { sharedBuild } from "./shared-build.js";
import { LOCAL_EMULATOR_STATE_PREFIX, LOCAL_SIMULATOR_STATE_PREFIX, assertLocalEmulatorCompatible, createLocalSimulator, startLocalEmulator, waitForLocalEmulator, waitForLocalSimulator, deleteLocalEmulator,
import { LOCAL_EMULATOR_STATE_PREFIX, LOCAL_SIMULATOR_STATE_PREFIX, agentDeviceClaims, assertLocalEmulatorCompatible, createLocalSimulator, startLocalEmulator, waitForLocalEmulator, waitForLocalSimulator, deleteLocalEmulator,
deleteLocalSimulator, hostEmulatorAbi, installOnLocalEmulator, installOnLocalSimulator, approveLinkScheme, localPoseShown,
setLocalEmulatorAppearance, setLocalSimulatorAppearance, startLocalArgentServer, stopLocalAgentDevice, stopLocalArgentServer, type LocalArgentServer } from "./local-simulator.js";
import { ensureArgentToolServer } from "../shared/controller-cache.js";
@@ -98,10 +98,6 @@ export type CiMapResult = {
};
function parseJsonOrUndefined(text: string): unknown {
try { return JSON.parse(text); } catch { return undefined; }
}
/**
* Minutes an EAS session is asked for beyond the capture's own budget. The session starts beside the repack, and the
* capture deadline only starts once the app is installed, so the session also has to outlast a slow or queued hosted
@@ -475,13 +471,7 @@ async function executeCiMap(options: CiMapOptions, setStage: (stage: string) =>
say(`starting a local ${options.device} emulator`);
localEmulatorStart = await startLocalEmulator({ runner, device: options.device, name: `explore-${side}-${options.gitCommit.slice(0, 8)}`, directory: simulatorProject,
cwd: simulatorProject, environment: restrictedEnvironment,
// agent-device refuses a serial an interrupted run still claims, even one it reports as stale, so that port is skipped.
...(controller === "agent-device" ? { claimed: (serial: string) => {
const status = runner({ command: options.agentDeviceCommand, args: ["device", "status", "--platform", "android", "--serial", serial, "--json"],
cwd: simulatorProject, environment: restrictedEnvironment, timeoutMs: 30_000, allowFailure: true });
const data = (parseJsonOrUndefined(status.stdout) as { data?: { claims?: unknown[]; hiddenStaleClaims?: number } } | undefined)?.data;
return (data?.claims?.length ?? 0) + (data?.hiddenStaleClaims ?? 0) > 0;
} } : {}),
...(controller === "agent-device" ? { claimed: (serial: string) => agentDeviceClaims({ runner, agentDevice: options.agentDeviceCommand, serial, cwd: simulatorProject, environment: restrictedEnvironment }) } : {}),
// Recorded as soon as it starts, so an interrupted review stops it too.
started: async serial => {
localEmulator = serial;
+2 -2
View File
@@ -90,9 +90,9 @@ const flagKinds: Record<string, Record<string, "value" | "boolean">> = {
"explore:binary": { archive:"value", output:"value", platform:"value", controller:"value", device:"value", "bundle-id":"value", "launch-url":"value", hints:"value", "allow-destructive":"value",
"explore:binary": { archive:"value", output:"value", platform:"value", controller:"value", simulator:"value", device:"value", "bundle-id":"value", "launch-url":"value", hints:"value", "allow-destructive":"value",
"max-duration-minutes":"value", "max-screens":"value", "max-actions":"value", provider:"value", model:"value", owner:"value", project:"value",
upload:"boolean", "no-upload":"boolean", "expires-in":"value", alias:"value" },
upload:"boolean", "no-upload":"boolean", "expires-in":"value", alias:"value", "no-open":"boolean", port:"value" },
"explore:session": { session:"value", "env-file":"value", "bundle-id":"value", "app-name":"value", "app-version":"value", output:"value", platform:"value", controller:"value", device:"value", "launch-url":"value", hints:"value",
"allow-destructive":"value", "max-duration-minutes":"value", "max-screens":"value", "max-actions":"value", provider:"value", model:"value", owner:"value", project:"value",
upload:"boolean", "expires-in":"value", alias:"value", "no-open":"boolean", port:"value" },
+16 -6
View File
@@ -39,12 +39,17 @@ export async function exploreBinary(): Promise<void> {
const platform = flag("platform") === undefined ? undefined : oneOf("platform", ["ios", "android"] as const);
const provider = flag("provider") === undefined ? undefined : oneOf("provider", ["anthropic", "openai"] as const);
const controller = flag("controller") === undefined ? undefined : oneOf("controller", ["agent-device", "argent"] as const);
const simulator = flag("simulator") === undefined ? "eas" : oneOf("simulator", ["eas", "local"] as const);
// On EAS Simulator the report is hosted unless --no-upload; locally it opens in the Viewer unless --upload, so a
// local run needs no Expo account at all.
const upload = simulator === "local" ? has("upload") : !has("no-upload");
const port = flag("port") === undefined ? undefined : nonNegativeInteger("port");
const started = Date.now();
const say = (message: string) => { process.stderr.write(`[explore ${Math.round((Date.now() - started) / 1000)}s] ${message}\n`); };
const allowDestructive = flags("allow-destructive").flatMap(value => value.split(",")).map(value => value.trim()).filter(Boolean);
if (!has("no-upload")) await assertHostingExport();
if (upload) await assertHostingExport();
const result = await runBinaryExploration({
archive: resolve(process.cwd(), archive), output: outputDirectory, progress: say,
archive: resolve(process.cwd(), archive), output: outputDirectory, progress: say, simulator,
...(platform ? { platform } : {}), ...(provider ? { provider } : {}), ...(controller ? { controller } : {}),
...(flag("device") ? { device: flag("device")! } : {}), ...(flag("bundle-id") ? { bundleId: flag("bundle-id")! } : {}),
...(flag("launch-url") ? { launchUrl: flag("launch-url")! } : {}), ...(flag("hints") ? { hints: flag("hints")! } : {}),
@@ -58,15 +63,16 @@ export async function exploreBinary(): Promise<void> {
const report = await writeExplorationReport(result, join(outputDirectory, "report"));
await keepReports(async () => [{ platform: result.archive.platform, zip: await zipReportDirectory(report.reportPath) }]);
let hosted: Awaited<ReturnType<typeof deployExplorationReport>> | undefined;
if (!has("no-upload")) {
if (upload) {
say("hosting the report on EAS Hosting");
hosted = await deployExplorationReport({ reportPath: report.reportPath, platform: result.archive.platform, project: flag("project") ?? DEFAULT_BINARY_PROJECT, owner: result.session.owner, expiresAfterDays, ...stableLinkOptions() });
const owner = result.session.owner || flag("owner");
hosted = await deployExplorationReport({ reportPath: report.reportPath, platform: result.archive.platform, project: flag("project") ?? DEFAULT_BINARY_PROJECT, ...(owner ? { owner } : {}), expiresAfterDays, ...stableLinkOptions() });
say(`report: ${hosted.previewUrl}${hosted.expiresAt ? ` (expires ${hosted.expiresAt})` : ""}`);
}
const summary = { ok: true,
app: { name: result.archive.appName, bundleId: result.archive.bundleId, platform: result.archive.platform, ...(result.archive.version ? { version: result.archive.version } : {}) },
archive: { path: result.archive.path, kind: result.archive.kind, sha256: result.archive.sha256, bytes: result.archive.bytes },
device: { model: result.runtime.deviceModel, osVersion: result.runtime.osVersion }, session: { id: result.session.id, owner: result.session.owner, project: result.session.project },
device: { model: result.runtime.deviceModel, osVersion: result.runtime.osVersion }, session: { id: result.session.id, simulator, ...(simulator === "eas" ? { owner: result.session.owner, project: result.session.project } : {}) },
model: result.model, exploration: { stopReason: result.exploration.stopReason, ...(result.exploration.summary ? { summary: result.exploration.summary } : {}),
screens: result.exploration.screens, observations: result.exploration.observations, actions: result.exploration.actions, durationMs: result.durationMs },
map: result.mapPath, report: report.reportPath, ...(hosted ? { url: hosted.previewUrl, expiresAt: hosted.expiresAt, deploymentUrl: hosted.deploymentUrl, ...(hosted.aliasUrl ? { aliasUrl: hosted.aliasUrl } : {}) } : {}),
@@ -75,7 +81,11 @@ export async function exploreBinary(): Promise<void> {
console.log(`${summary.app.name} (${summary.app.bundleId}, ${summary.app.platform}) explored on ${summary.device.model} ${summary.device.osVersion}: ${summary.exploration.screens} screens, ${summary.exploration.actions} actions, stopped because ${summary.exploration.stopReason}.`);
if (hosted) console.log(`Hosted report: ${hosted.previewUrl}${hosted.expiresAt ? ` (expires ${hosted.expiresAt})` : ""}`);
console.log(`Local report: ${report.reportPath}`);
console.log(`Next: ${summary.next.join(" | ")}`);
if (simulator === "eas" || hosted || has("no-open")) { console.log(`Next: ${summary.next.join(" | ")}`); return; }
const viewer = await startViewerServer({ root: outputDirectory, ...(port === undefined ? {} : { port }) });
console.log(`Viewer: ${viewer.url} (Ctrl+C to stop)`);
openViewerInBrowser(viewer.url);
await waitForViewerShutdown(viewer);
}
/** `--env-file` wins; otherwise the `.env.eas-simulator` that `eas simulator:start --out-config-type dotenv` wrote in the current directory, unless the environment already carries a session. */
+9 -6
View File
@@ -19,8 +19,8 @@ const GET_STARTED = `Get started (run these in the Expo app's directory):
(captures the committed HEAD like the baseline workflow, with your eas login, and reports each route)
capture [--base REF] [--head REF | --committed] [--target ID | --platform ios|android] [--parallel N] [--scope all|changes]
[--controller agent-device|argent] [--simulator eas|local] [--repack eas|local] [--upload [--expires-in 7d|30d|never]] [--no-open] [--no-cache] [--output DIRECTORY] [--project DIRECTORY]
(on this machine: checks out both commits in temporary worktrees, captures them with your eas login and your
local claude or codex login, and opens the report; the head is HEAD with your uncommitted changes, as eas build
(on this machine: checks out both commits in temporary worktrees, captures them with your eas login (in agent
mode, also your local claude or codex login), and opens the report; the head is HEAD with your uncommitted changes, as eas build
uploads them (--committed, or cli.requireCommit in eas.json, leaves them out); the base defaults to where HEAD
left the default branch;
without .expo-explore/config.yaml it uses the defaults setup would write;
@@ -43,19 +43,22 @@ const GET_STARTED = `Get started (run these in the Expo app's directory):
(hosts reports already on this machine on EAS Hosting behind one link, with your eas login: one per platform
of the same review, from capture:review --output or a downloaded explore-review artifact, or one explore:binary report)`;
const BINARY = `Prebuilt apps (no source needed; runs on EAS Simulator with your local claude or codex login):
explore:binary <app.apk | app.app | app.tar.gz | simulator .ipa> [--output DIRECTORY] [--no-upload] [--json]
const BINARY = `Prebuilt apps (no source needed; runs with your local claude or codex login):
explore:binary <app.apk | app.app | app.tar.gz | simulator .ipa> [--output DIRECTORY] [--simulator eas|local] [--no-upload] [--json]
(installs the archive in a fresh EAS Simulator session, lets the model explore it, writes a map
and report under --output, hosts the report on EAS Hosting, and prints the link)
--simulator local runs it on a simulator or emulator it starts on this machine instead, with no EAS login,
and opens the report in the local Viewer (--upload hosts it instead, --no-open skips the Viewer, --port N);
iOS needs Xcode, Android the SDK (ANDROID_HOME) with platform-tools, cmdline-tools, the emulator and a system image for this Mac
Flags, all optional: --device NAME --bundle-id ID --launch-url URL --hints "what to focus on"
--controller agent-device|argent (the session's device controller; agent-device by default)
--controller agent-device|argent (the device controller; agent-device by default)
(a sign-in in --hints is written as {{secret:NAME}}: Explore types the value of the environment variable NAME,
and the model and the report only see the reference)
--max-duration-minutes N --max-screens N --max-actions N --provider anthropic|openai --model ID
--owner ACCOUNT --project SLUG --expires-in 7d|30d|never --allow-destructive "Label,Other label"
--alias NAME (one EAS Hosting host per app that always serves the newest run; with EXPLORE_PREVIEW_TOKEN
set to a fixed 32-character token, the hosted link is the same every run)
(iOS needs a simulator build; a device .ipa and Android .aab are rejected before any session starts)
(iOS needs a simulator build; a device .ipa and Android .aab are rejected before any device starts)
explore:session [SESSION_ID] [--bundle-id ID] [--output DIRECTORY] [--upload] [--json]
(explores an app that is already installed on an EAS Simulator session you started — run it where
"eas simulator:start --type agent-device --out-config-type dotenv" (or --type argent) wrote .env.eas-simulator,
+1 -1
View File
@@ -577,7 +577,7 @@ export async function deployExplorationReport(options: ExplorationDeployOptions)
message: "An EAS login is required to host the report", nextAction: "Run `eas login` or set EXPO_TOKEN, then rerun with --upload." });
if (!/^[a-z0-9][a-z0-9-]{0,38}[a-z0-9]$/.test(options.project)) throw previewError("PREVIEW_PROJECT_INVALID", `Invalid EAS project slug: ${options.project}`, "Pass --project as a lowercase slug such as expo-explore.");
if (options.alias !== undefined && !HOSTING_ALIAS.test(options.alias)) throw previewError("PREVIEW_ALIAS_INVALID", `Invalid EAS Hosting alias: ${options.alias}`, "Pass --alias as a lowercase slug such as bluesky.");
if (options.token !== undefined && !HOSTED_TOKEN.test(options.token)) throw previewError("PREVIEW_TOKEN_INVALID", "EXPLORE_PREVIEW_TOKEN must be 32 characters of a-z and 2-7.", "Generate one with `openssl rand 20 | base32 | tr A-Z a-z` and set EXPLORE_PREVIEW_TOKEN, or unset it for a random token.");
if (options.token !== undefined && !HOSTED_TOKEN.test(options.token)) throw previewError("PREVIEW_TOKEN_INVALID", "EXPLORE_PREVIEW_TOKEN must be 32 characters of a-z and 2-7.", "Generate one with `LC_ALL=C tr -dc a-z2-7 </dev/urandom | head -c 32` and set EXPLORE_PREVIEW_TOKEN, or unset it for a random token.");
const exportPath = resolve(options.exportPath ?? PREVIEW_EXPORT_PATH);
await assertExport(exportPath);
let zip: Uint8Array;
+4 -1
View File
@@ -22,7 +22,7 @@ function fail(args) {
test("explore:binary is in the short help with its archive kinds and --no-upload", () => {
const help = spawnSync(process.execPath, [cli, "help"], { cwd: root, encoding: "utf8" });
assert.equal(help.status, 0, help.stderr);
assert.match(help.stdout, /explore:binary <app\.apk \| app\.app \| app\.tar\.gz \| simulator \.ipa>.*--no-upload/);
assert.match(help.stdout, /explore:binary <app\.apk \| app\.app \| app\.tar\.gz \| simulator \.ipa>.*--simulator eas\|local.*--no-upload/);
});
test("explore:binary validates its arguments before touching EAS or a model", () => {
@@ -30,6 +30,9 @@ test("explore:binary validates its arguments before touching EAS or a model", ()
assert.match(fail(["./app.apk", "--upload", "--no-upload"]), /either --upload or --no-upload/);
assert.match(fail(["./app.apk", "--expires-in", "400d"]), /--expires-in must be 1d to 90d, or never/);
assert.match(fail(["./app.apk", "--platform", "web"]), /platform/);
assert.match(fail(["./app.apk", "--simulator", "cloud"]), /simulator/);
// A local run needs no EAS login, so the archive is the first thing it checks.
assert.match(fail(["./does-not-exist.apk", "--simulator", "local"]), /Archive not found: .*does-not-exist\.apk/);
assert.match(fail(["./does-not-exist.apk", "--no-upload"]), /Archive not found: .*does-not-exist\.apk/);
assert.match(fail(["./package.json", "--no-upload"]), /Unsupported archive: package\.json/);
const human = spawnSync(process.execPath, [cli, "explore:binary", "./does-not-exist.apk", "--no-upload"], { cwd: root, encoding: "utf8", env: { ...process.env, EXPO_TOKEN: "", EXPO_SESSION_SECRET: "" } });
+87
View File
@@ -0,0 +1,87 @@
import assert from "node:assert/strict";
import { writeFileSync } from "node:fs";
import { mkdir, mkdtemp, readdir, rm, writeFile } from "node:fs/promises";
import { tmpdir } from "node:os";
import { join } from "node:path";
import test from "node:test";
import { tsImport } from "tsx/esm/api";
const { runBinaryExploration } = await tsImport("../../src/binary/run.ts", import.meta.url);
const { controllerVersion } = await tsImport("../../src/capture/simulator.ts", import.meta.url);
const { CONTROLLER_SESSION } = await tsImport("../../src/capture/common.ts", import.meta.url);
const UDID = "0A1B2C3D-4E5F-4A6B-8C7D-9E0F1A2B3C4D";
const PNG = Buffer.from("iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNkYPhfDwAChwGA60e6kgAAAABJRU5ErkJggg==", "base64");
const DEVICE_TYPE = "com.apple.CoreSimulator.SimDeviceType.iPhone-17-Pro";
async function simulatorApp(root) {
const app = join(root, "Example.app");
await mkdir(app, { recursive: true });
await writeFile(join(app, "Info.plist"), `<?xml version="1.0" encoding="UTF-8"?>
<plist version="1.0"><dict>
<key>CFBundleIdentifier</key><string>dev.example.app</string>
<key>CFBundleDisplayName</key><string>Example</string>
<key>DTPlatformName</key><string>iphonesimulator</string>
<key>CFBundleURLTypes</key><array><dict><key>CFBundleURLSchemes</key><array><string>example</string></array></dict></array>
</dict></plist>`);
await writeFile(join(app, "Example"), "binary");
return app;
}
/** simctl and agent-device on a Mac with one iPhone 17 Pro runtime; the app shows a single screen. */
function localRunner() {
const requests = [];
const ok = data => ({ stdout: JSON.stringify({ success: true, data }), stderr: "" });
const runner = request => {
requests.push(request);
const [first, second, third] = request.args;
if (request.command === "xcrun") {
if (second === "list" && third === "devicetypes") return { stdout: JSON.stringify({ devicetypes: [{ name: "iPhone 17 Pro", identifier: DEVICE_TYPE }] }), stderr: "" };
if (second === "list" && third === "runtimes") return { stdout: JSON.stringify({ runtimes: [{ isAvailable: true, platform: "iOS", identifier: "com.apple.CoreSimulator.SimRuntime.iOS-27-0",
version: "27.0", supportedDeviceTypes: [{ identifier: DEVICE_TYPE }] }] }), stderr: "" };
if (second === "create") return { stdout: `${UDID}\n`, stderr: "" };
return { stdout: "", stderr: "" };
}
if (first === "--version") return { stdout: `${controllerVersion("agent-device")}\n`, stderr: "" };
if (first === "open") return ok({ session: CONTROLLER_SESSION, appBundleId: "dev.example.app", device: { platform: "ios", target: "mobile", name: "Explore iPhone 17 Pro", id: UDID },
identifiers: { session: CONTROLLER_SESSION, appBundleId: "dev.example.app", deviceName: "Explore iPhone 17 Pro", deviceId: UDID } });
if (first === "snapshot") return ok({ nodes: [{ type: "StaticText", label: "Welcome" }, { type: "Button", label: "Settings" }] });
if (first === "appstate") return ok({ platform: "ios", appBundleId: "dev.example.app", state: "runningForeground", surface: "app" });
if (first === "screenshot") { writeFileSync(request.args[1], PNG); return ok({}); }
return ok({});
};
return { requests, runner };
}
test("explore:binary --simulator local runs on a simulator it creates, with no EAS login, and deletes it afterwards", async t => {
const root = await mkdtemp(join(tmpdir(), "explore-binary-local-"));
t.after(() => rm(root, { recursive: true, force: true }));
const { requests, runner } = localRunner();
const session = { provider: "anthropic", model: "test", async preflight() {}, async run() { return { stopReason: "done" }; } };
const result = await runBinaryExploration({ archive: await simulatorApp(root), output: join(root, "out"), simulator: "local", provider: "anthropic", runner,
createSession: () => session, environment: { PATH: process.env.PATH, HOME: root, ANTHROPIC_API_KEY: "key" } });
// No EAS: no login check, project, or session.
assert.equal(requests.some(request => request.args.some(arg => /^(whoami|init|simulator:)/.test(arg))), false);
const simctl = requests.filter(request => request.command === "xcrun").map(request => request.args.slice(1).join(" "));
assert.ok(simctl.some(command => command.startsWith(`create Explore iPhone 17 Pro (binary `)));
assert.ok(simctl.includes(`spawn ${UDID} defaults write com.apple.launchservices.schemeapproval com.apple.CoreSimulator.CoreSimulatorBridge-->example -string dev.example.app`),
"the app's own scheme opens without a prompt");
assert.ok(simctl.some(command => command.startsWith(`install ${UDID} `)), "simctl installs the app");
assert.equal(simctl.at(-1), `delete ${UDID}`, "and the simulator goes last");
// agent-device runs its own daemon in the run's directory and is pointed at the simulator by its UDID.
const controller = requests.filter(request => request.command !== "xcrun" && request.args[0] !== "--version" && request.args[0] !== "close");
assert.ok(controller.length > 0);
for (const request of controller) {
assert.deepEqual(request.args.slice(request.args.indexOf("--udid"), request.args.indexOf("--udid") + 2), ["--udid", UDID]);
assert.match(request.environment.AGENT_DEVICE_STATE_DIR, /agent-device-state$/);
assert.equal(request.environment.AGENT_DEVICE_DAEMON_BASE_URL, undefined);
}
assert.equal(controller.some(request => request.args[0] === "install"), false, "the controller does not install it again");
assert.equal(result.runtime.provider, "local-simulator");
assert.equal(result.runtime.osVersion, "27.0");
assert.equal(result.session.id, `local-${UDID}`);
assert.deepEqual(await readdir(join(root, ".cache", "expo-explore", "local-captures")), [], "nothing is left for a sweep");
});