feat: enable Computer Use with macOS + Windows + Linux support

Phase 1: Replace @ant/computer-use-mcp stub (12 files, 6517 lines).

Phase 2: Remove 8 macOS-only guards in src/:
- main.tsx: remove getPlatform()==='macos' check
- swiftLoader.ts: remove darwin-only throw
- executor.ts: extend platform guard, clipboard dispatch, paste key
- drainRunLoop.ts: skip CFRunLoop pump on non-darwin
- escHotkey.ts: non-darwin returns false (Ctrl+C fallback)
- hostAdapter.ts: non-darwin permissions granted
- common.ts: dynamic platform + screenshotFiltering
- gates.ts: enabled:true, subscription check removed

Phase 3: Add Linux backends (xdotool/scrot/xrandr/wmctrl):
- computer-use-input/backends/linux.ts (173 lines)
- computer-use-swift/backends/linux.ts (278 lines)

Verified on Windows x64: mouse, screenshot, displays, foreground app.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
unraid
2026-04-03 22:33:00 +08:00
parent 465e9f01c6
commit e3264a1691
34 changed files with 8291 additions and 750 deletions

View File

@@ -52,8 +52,8 @@ export function getTerminalBundleId(): string | null {
* takes this shape (no `hostBundleId`, no `teachMode`).
*/
export const CLI_CU_CAPABILITIES = {
screenshotFiltering: 'native' as const,
platform: 'darwin' as const,
screenshotFiltering: (process.platform === 'darwin' ? 'native' : 'none') as any,
platform: (process.platform === 'win32' ? 'windows' : process.platform === 'linux' ? 'linux' : 'darwin') as any,
}
export function isComputerUseMCPServer(name: string): boolean {

View File

@@ -59,6 +59,7 @@ export const releasePump = release
* concurrent drainRunLoop() calls share one setInterval.
*/
export async function drainRunLoop<T>(fn: () => Promise<T>): Promise<T> {
if (process.platform !== 'darwin') return fn()
retain()
let timer: ReturnType<typeof setTimeout> | undefined
try {

View File

@@ -23,6 +23,7 @@ import { requireComputerUseSwift } from './swiftLoader.js'
let registered = false
export function registerEscHotkey(onEscape: () => void): boolean {
if (process.platform !== 'darwin') return false
if (registered) return true
const cu = requireComputerUseSwift()
if (!(cu as any).hotkey.registerEscape(onEscape)) {

View File

@@ -68,6 +68,24 @@ function computeTargetDims(
}
async function readClipboardViaPbpaste(): Promise<string> {
if (process.platform === 'win32') {
const { stdout, code } = await execFileNoThrow('powershell', ['-NoProfile', '-Command', 'Get-Clipboard'], {
useCwd: false,
})
if (code !== 0) {
throw new Error(`PowerShell Get-Clipboard exited with code ${code}`)
}
return stdout
}
if (process.platform === 'linux') {
const { stdout, code } = await execFileNoThrow('xclip', ['-selection', 'clipboard', '-o'], {
useCwd: false,
})
if (code !== 0) {
throw new Error(`xclip exited with code ${code}`)
}
return stdout
}
const { stdout, code } = await execFileNoThrow('pbpaste', [], {
useCwd: false,
})
@@ -78,6 +96,25 @@ async function readClipboardViaPbpaste(): Promise<string> {
}
async function writeClipboardViaPbcopy(text: string): Promise<void> {
if (process.platform === 'win32') {
const { code } = await execFileNoThrow('powershell', ['-NoProfile', '-Command', `Set-Clipboard -Value '${text.replace(/'/g, "''")}'`], {
useCwd: false,
})
if (code !== 0) {
throw new Error(`PowerShell Set-Clipboard exited with code ${code}`)
}
return
}
if (process.platform === 'linux') {
const { code } = await execFileNoThrow('xclip', ['-selection', 'clipboard'], {
input: text,
useCwd: false,
})
if (code !== 0) {
throw new Error(`xclip exited with code ${code}`)
}
return
}
const { code } = await execFileNoThrow('pbcopy', [], {
input: text,
useCwd: false,
@@ -192,7 +229,7 @@ async function typeViaClipboard(input: Input, text: string): Promise<void> {
if ((await readClipboardViaPbpaste()) !== text) {
throw new Error('Clipboard write did not round-trip.')
}
await input.keys(['command', 'v'])
await input.keys([process.platform === 'darwin' ? 'command' : 'ctrl', 'v'])
await sleep(100)
} finally {
if (typeof saved === 'string') {
@@ -260,9 +297,9 @@ export function createCliExecutor(opts: {
getMouseAnimationEnabled: () => boolean
getHideBeforeActionEnabled: () => boolean
}): ComputerExecutor {
if (process.platform !== 'darwin') {
if (process.platform !== 'darwin' && process.platform !== 'win32' && process.platform !== 'linux') {
throw new Error(
`createCliExecutor called on ${process.platform}. Computer control is macOS-only.`,
`createCliExecutor called on ${process.platform}. Computer control requires macOS, Windows, or Linux.`,
)
}

View File

@@ -10,7 +10,7 @@ type ChicagoConfig = CuSubGates & {
}
const DEFAULTS: ChicagoConfig = {
enabled: false,
enabled: true,
pixelValidation: false,
clipboardPasteMultiline: true,
mouseAnimation: true,
@@ -37,9 +37,7 @@ function readConfig(): ChicagoConfig {
// regardless of subscription tier — not all ants are max/pro, and per
// CLAUDE.md:281, USER_TYPE !== 'ant' branches get zero antfooding.
function hasRequiredSubscription(): boolean {
if (process.env.USER_TYPE === 'ant') return true
const tier = getSubscriptionType()
return tier === 'max' || tier === 'pro'
return true
}
export function getChicagoEnabled(): boolean {

View File

@@ -45,6 +45,7 @@ export function getComputerUseHostAdapter(): ComputerUseHostAdapter {
getHideBeforeActionEnabled: () => getChicagoSubGates().hideBeforeAction,
}),
ensureOsPermissions: async () => {
if (process.platform !== 'darwin') return { granted: true }
const cu = requireComputerUseSwift()
const accessibility = (cu as any).tcc.checkAccessibility()
const screenRecording = (cu as any).tcc.checkScreenRecording()

View File

@@ -13,9 +13,6 @@ let cached: ComputerUseAPI | undefined
* these in drainRunLoop().
*/
export function requireComputerUseSwift(): ComputerUseAPI {
if (process.platform !== 'darwin') {
throw new Error('@ant/computer-use-swift is macOS-only')
}
// eslint-disable-next-line @typescript-eslint/no-require-imports
return (cached ??= require('@ant/computer-use-swift') as ComputerUseAPI)
}