diff --git a/.claude-plugin/marketplace.json b/.claude-plugin/marketplace.json index 8840b74..4bf5c64 100644 --- a/.claude-plugin/marketplace.json +++ b/.claude-plugin/marketplace.json @@ -844,7 +844,7 @@ "description": "Programmatic screenshot capture on macOS. Get window IDs via Swift CGWindowListCopyWindowInfo, capture specific windows with screencapture -l, and control application windows via AppleScript. Supports multi-shot workflows for capturing different sections of the same window. Use when taking automated screenshots, capturing application windows, or creating visual documentation", "source": "./", "strict": false, - "version": "1.0.0", + "version": "1.0.1", "category": "utilities", "keywords": [ "screenshot", @@ -947,4 +947,4 @@ ] } ] -} \ No newline at end of file +} diff --git a/capture-screen/SKILL.md b/capture-screen/SKILL.md index ab88d26..8cf693b 100644 --- a/capture-screen/SKILL.md +++ b/capture-screen/SKILL.md @@ -227,6 +227,65 @@ These methods were tested and confirmed to fail on macOS: | Python `import Quartz` (PyObjC) | `ModuleNotFoundError` | PyObjC not installed in system Python; don't attempt to install it — use Swift instead | | `osascript` window id | Wrong format | Returns AppleScript window index, not CGWindowID needed by `screencapture -l` | +## Permission Troubleshooting + +`swift scripts/get_window_id.swift` reads on-screen windows via CoreGraphics, so it needs Screen Recording permission on macOS. + +Use this order: + +1. Confirm trigger +2. Confirm target identity +3. Add/enable exact app in Settings + +If the command fails with `ERROR: Failed to enumerate windows`, do this: + +```bash +open "x-apple.systempreferences:com.apple.preference.security?Privacy_ScreenCapture" +``` + +Or print the same checklist directly from the script: + +```bash +swift scripts/get_window_id.swift --permission-hint screen +swift scripts/get_window_id.swift --permission-hint microphone +``` + +Then: + +1. In Privacy & Security → Screen Recording, enable the target app. +2. If your app is missing from the list: + - Ensure you granted permission to the real app bundle (not `swift` / terminal helpers). + - For CLI tools, build/run as a packaged `.app` during permission verification. + - Click `+` and add the `.app` manually from `/Applications`. +3. Re-run the command after restarting the app. +4. If this is a CLI workflow, also check whether the launcher is a helper binary: + - In most cases the entry shown in TCC is the helper process (`swift`, `Terminal`, `iTerm`, etc.), not the business app. + - Permission still works after helper-level grant, but it is not ideal for final UX. + +For mic-access-related prompts, use the same pattern with the microphone pane: + +```bash +open "x-apple.systempreferences:com.apple.preference.security?Privacy_Microphone" +``` + +The same rule still applies: the system can only show permissions for a concrete `.app` bundle. If the request is made by a helper binary, the settings list can be misleading or empty for your product app. + +### Quick Check Template + +```text +1) Error: permission denied +2) Open target pane +3) Verify identity shown by OS = identity you granted +4) If not matched, use the script-reported candidate identities and grant the launcher process +5) Reopen/restart and verify +``` + +For production apps, avoid requesting permissions via `swift`/`python` entry points; always route permission checks in the packaged app process so users only see one target. + +If you maintain another macOS permission-related flow, reuse this standardized triage template: + +- [permission-triage-template.md](references/permission-triage-template.md) + ## Supported Applications | Application | Window ID | AppleScript Control | Notes | diff --git a/capture-screen/references/permission-triage-template.md b/capture-screen/references/permission-triage-template.md new file mode 100644 index 0000000..16c3919 --- /dev/null +++ b/capture-screen/references/permission-triage-template.md @@ -0,0 +1,42 @@ +# macOS 权限排障模板(Screen Recording / 麦克风) + +## 排障目标 +- 在系统设置里找不到目标应用 +- 权限拒绝但设置项看起来已打开 +- 通过终端/脚本入口触发时,用户不知道该给谁授权 + +## 标准排查顺序(必须按序执行) + +1. 确认触发点 + - 明确是哪个权限被拒绝(Screen Recording / 麦克风)。 +2. 确认 TCC 实体 + - 不是脚本文件名。 + - 先确认“当前触发进程”与“最终应用体”是否一致。 + - 关注脚本输出里的候选身份列表(invoker/runtime)并逐项核验。 +3. 确认设置面板 + - 直接跳转到对应隐私面板 + - 允许该进程/应用 + - 重启进程后复验 + +## 通用动作模板 + +```bash +# Screen Recording +open "x-apple.systempreferences:com.apple.preference.security?Privacy_ScreenCapture" + +# Microphone +open "x-apple.systempreferences:com.apple.preference.security?Privacy_Microphone" +``` + +## 不在列表时处理 + +- 优先确认请求来自真实 .app Bundle(签名、打包) +- 如果当前为 CLI/脚本入口,先给宿主进程授权(Terminal/iTerm/swift/python) +- 在设置面板点击 `+` 手工添加目标 `.app` +- 变更后退出并重启应用,重新测试 + +## 验收标准(用户侧) + +- 用户能看到一条明确的“应授权对象” +- 错误提示中有“找不到对象时下一步该做什么” +- 无需反复猜测在设置里要点击什么 diff --git a/capture-screen/scripts/get_window_id.swift b/capture-screen/scripts/get_window_id.swift index c19ef7f..8ccef14 100755 --- a/capture-screen/scripts/get_window_id.swift +++ b/capture-screen/scripts/get_window_id.swift @@ -4,9 +4,11 @@ // Enumerate on-screen windows and print their Window IDs. // // Usage: -// swift get_window_id.swift # List all windows -// swift get_window_id.swift Excel # Filter by keyword -// swift get_window_id.swift "Chrome" # Filter by app name +// swift scripts/get_window_id.swift # List all windows +// swift scripts/get_window_id.swift Excel # Filter by keyword +// swift scripts/get_window_id.swift "Chrome" # Filter by app name +// swift scripts/get_window_id.swift --permission-hint screen # Print Screen Recording triage +// swift scripts/get_window_id.swift --permission-hint microphone # Print Microphone triage // // Output format: // WID=12345 | App=Microsoft Excel | Title=workbook.xlsx @@ -15,10 +17,192 @@ // import CoreGraphics +import Foundation -let keyword = CommandLine.arguments.count > 1 - ? CommandLine.arguments[1] - : "" +enum PermissionKind: String { + case screen + case microphone +} + +let invocationPath = (CommandLine.arguments.first ?? "") +let invokerName = URL(fileURLWithPath: invocationPath).lastPathComponent +let runtimeProcessName = ProcessInfo.processInfo.processName +let invokerIsBundle = invocationPath.hasSuffix(".app") || invocationPath.contains(".app/") +let scriptPath: String? = invocationPath.hasSuffix(".swift") ? invocationPath : nil +let helperBinaries: Set = [ + "swift", + "swift-frontend", + "python", + "python3", + "node", + "uv", + "npm", + "bun", + "pnpm", + "yarn", + "bash", + "zsh", + "sh", + "osascript", + "Terminal", + "iTerm2", + "iTerm" +] + +let invokerCandidates: [String] = { + var candidates = [String]() + var seen = Set() + func append(_ value: String) { + guard !value.isEmpty, !seen.contains(value) else { return } + seen.insert(value) + candidates.append(value) + } + if let scriptPath = scriptPath { + append(scriptPath) + } + if !invokerName.isEmpty { + append(invokerName) + } + if !runtimeProcessName.isEmpty && runtimeProcessName != invokerName { + append(runtimeProcessName) + } + return candidates +}() + +let args = Array(CommandLine.arguments.dropFirst()) + +var permissionHintTarget: PermissionKind? +var keyword = "" +var expectPermissionTarget = false + +func printUsage() { + fputs("Usage:\n", stderr) + fputs(" swift scripts/get_window_id.swift [keyword]\n", stderr) + fputs(" swift scripts/get_window_id.swift --permission-hint [screen|microphone]\n", stderr) + fputs("\n", stderr) + fputs("Options:\n", stderr) + fputs(" --permission-hint [screen|microphone] Print permission triage instructions\n", stderr) + fputs(" -h, --help Show this help\n", stderr) + fputs("\n", stderr) + fputs("Examples:\n", stderr) + fputs(" swift scripts/get_window_id.swift Excel\n", stderr) + fputs(" swift scripts/get_window_id.swift --permission-hint screen\n", stderr) + fputs(" swift scripts/get_window_id.swift --permission-hint microphone\n", stderr) +} + +for arg in args { + if expectPermissionTarget { + if let kind = PermissionKind(rawValue: arg.lowercased()) { + permissionHintTarget = kind + expectPermissionTarget = false + continue + } + fputs("Unknown permission target: \(arg)\n", stderr) + printUsage() + exit(2) + } + + if arg == "-h" || arg == "--help" { + printUsage() + exit(0) + } else if arg == "--permission-hint" { + permissionHintTarget = .screen + expectPermissionTarget = true + } else if arg.hasPrefix("--permission-hint=") { + let target = String(arg.dropFirst("--permission-hint=".count)).lowercased() + guard let kind = PermissionKind(rawValue: target) else { + fputs("Unknown permission target: \(target)\n", stderr) + printUsage() + exit(2) + } + permissionHintTarget = kind + } else if arg == "--permission-hint-screen" { + permissionHintTarget = .screen + } else if arg == "--permission-hint-microphone" || arg == "--permission-hint-mic" { + permissionHintTarget = .microphone + } else if arg.hasPrefix("-") { + fputs("Unknown option: \(arg)\n", stderr) + printUsage() + exit(2) + } else if keyword.isEmpty { + keyword = arg + } +} + +if expectPermissionTarget { + fputs("Missing permission hint target after --permission-hint.\n", stderr) + printUsage() + exit(2) +} + +if let kind = permissionHintTarget { + switch kind { + case .screen: + fputs("Screen Recording permission required.\n", stderr) + printCommonPermissionHint( + pane: "Privacy_ScreenCapture", + label: "Screen Recording" + ) + printPermissionContextHint() + case .microphone: + fputs("Microphone permission required.\n", stderr) + printCommonPermissionHint( + pane: "Privacy_Microphone", + label: "Microphone" + ) + printPermissionContextHint() + } + exit(0) +} + +func printCommonPermissionHint(pane: String, label: String, missing: Bool = true) { + let openCommand = "x-apple.systempreferences:com.apple.preference.security?\(pane)" + fputs("Troubleshooting:\n", stderr) + fputs(" - Open Settings: `open \"\(openCommand)\"`\n", stderr) + fputs(" - In Privacy & Security → \(label), enable the target application.\n", stderr) + if missing { + fputs(" - If the target app is not in the list:\n", stderr) + fputs(" - Granting happens by real .app bundle, not by helper/terminal scripts.\n", stderr) + fputs(" - For CLI workflows, grant to the host app you launch from (Terminal, iTerm, iTerm2, Swift, etc.) if no dedicated .app exists yet.\n", stderr) + fputs(" - Click `+` and add the actual `.app` from `/Applications`.\n", stderr) + } + fputs(" - If permission status does not refresh, quit/reopen terminal/app and retry.\n", stderr) + if helperBinaries.contains(invokerName) || helperBinaries.contains(runtimeProcessName) { + fputs(" - Current launcher is a helper/runtime process (`\(runtimeProcessName)`) -> OS may show this entry instead of the tool name.\n", stderr) + } else if invokerIsBundle { + fputs(" - The launcher path looks like a bundled app, which is the preferred state for permissions.\n", stderr) + } + if let scriptPath = scriptPath { + fputs(" - Script entry: `\(scriptPath)`\n", stderr) + } +} + +func printPermissionContextHint() { + fputs(" - Invoker path: `\(invocationPath)`\n", stderr) + fputs(" - Runtime process: `\(runtimeProcessName)`\n", stderr) + if !invokerCandidates.isEmpty { + fputs(" - Candidate identities in System Settings:\n", stderr) + for identity in invokerCandidates { + fputs(" - \(identity)\n", stderr) + } + } + if invokerIsBundle { + fputs(" This looks like a bundled app path, so the setting should match the app identity.\n", stderr) + } else { + fputs(" If this is not your final app binary, permissions can be inconsistent.\n", stderr) + } + fputs(" - Recommended for production: keep permission requests inside your signed `.app` process.\n", stderr) + if let scriptPath = scriptPath { + fputs(" - Script entry currently used: `\(scriptPath)`.\n", stderr) + } +} + +func printScreenRecordingPermissionHint() { + fputs("Screen Recording permission required.\n", stderr) + fputs("Troubleshooting:\n", stderr) + printCommonPermissionHint(pane: "Privacy_ScreenCapture", label: "Screen Recording") + printPermissionContextHint() +} guard let windowList = CGWindowListCopyWindowInfo( .optionOnScreenOnly, kCGNullWindowID @@ -27,6 +211,7 @@ guard let windowList = CGWindowListCopyWindowInfo( fputs("Possible causes:\n", stderr) fputs(" - No applications with visible windows are running\n", stderr) fputs(" - Screen Recording permission not granted (System Settings → Privacy & Security → Screen Recording)\n", stderr) + printScreenRecordingPermissionHint() exit(1) }