SilbercueSwift — iOS MCP Server

io.github.Silbercue由 GitHub 账号 Silbercue 发布v3.6.3更新于 Oct 24, 2026

iOS MCP server: 49 tools free (58 with Pro), xcresult parsing, WDA UI automation, fast screenshots.

已验证STDIO仅桌面浏览器自动化开发者工具

概览

AI 生成的概览

本地 macOS MCP 服务器,让 AI 助手在模拟器中构建、测试、运行并自动化操作 iOS 应用,提供结构化 xcresult 结果和截图。

可以做什么
构建与运行应用:为模拟器构建、一次调用完成构建并运行、清理、发现项目、列出 scheme,并从 xcresult 返回结构化错误。测试与诊断:运行测试、列出失败用例及 file:line 和失败截图、读取按文件的代码覆盖率。管理模拟器:列出、启动、关闭、安装、启动应用、终止、克隆、抹除、删除、旋转和查看详情。自动化 UI:查找元素、点击、输入、滑动、缩放、拖放、导航、处理弹窗、读取视图层级并执行多步计划。采集输出:截图、带主题过滤的 os_log 日志捕获、应用控制台输出以及等待日志模式。检查质量与版本控制:视觉回归、多设备、无障碍与本地化检查,以及 Git 状态、差异、日志、提交和分支。
适用场景
当助手需要在 Mac 上处理 iOS 应用时使用:在模拟器中构建和运行、运行测试并读取结构化失败信息、通过无障碍元素驱动 UI,或检查截图、日志、覆盖率、无障碍和本地化。它仅支持 iOS,不支持 Android、watchOS、tvOS 或 visionOS。
运行要求
在 macOS 13 或更高版本本地运行,需安装 Xcode 15 或更高版本(用于 xcresulttool 和 simctl);仅从源码构建时需要 Swift 6.0 或更高版本。UI 自动化工具需要在模拟器上安装 WebDriverAgent。可通过 Homebrew 安装或从源码构建,并以完整二进制路径配置为 stdio 服务器。未声明 API 密钥或环境变量。
安装前请注意
Git 工具包含写操作(git_commit、git_branch),模拟器工具可抹除或删除模拟器并终止应用。UI 自动化会在模拟器上点击、输入并关闭系统权限弹窗。Pro 功能需要付费许可证(每月 12 欧元),并用许可证密钥激活。README 称其完全在 Mac 上运行且不收集用户数据。

安装

在 SourceWeft 中

  1. 打开 控制台中的 SilbercueSwift — iOS MCP Server,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

SilbercueSwift

[GitHub Release] [Free — 49 tools] [Pro available] [License: MIT] [MCP Registry] [Platform] [Swift 6.0] [SilbercueSwift MCP server]

[Capabilities covered, of 19 compared: SilbercueSwift 18, XcodeBuildMCP 6, Appium-MCP 4, iosef 1]

The fastest, most complete MCP server for iOS development. One Swift binary, 58 tools, zero dependencies. SilbercueSwift has the most complete toolset of any alternative out there.

Built for Claude Code, Cursor, and any MCP-compatible AI agent.

Looking for an alternative to existing iOS MCP servers? SilbercueSwift covers the full feature set of XcodeBuildMCP, Appium-MCP, and iosef in a single binary — plus xcresult parsing, UI automation, code coverage, and up to 75x faster screenshots. See comparison below.

Why SilbercueSwift?

Every iOS MCP server has the same problem: raw xcodebuild output is useless for AI agents. 500 lines of build log, stderr noise mistaken for errors, no structured test results. Agents waste minutes parsing what a human sees in seconds.

SilbercueSwift fixes this. It parses .xcresult bundles — the same structured data Xcode uses internally — and returns exactly what the agent needs: pass/fail counts, failure messages with file:line, code coverage per file, and failure screenshots.

What you getXcodeBuildMCPAppium-MCPiosefSilbercueSwift
Screenshot latency~1127ms~77ms~83ms~316ms
([图片] ~15ms, 75x)
View hierarchy~259ms~938ms~44ms~31ms
([图片] ~5ms)
Find element—76ms50ms31ms
([图片] <1ms + auto-scroll)
Tap (coordinates)235ms470ms48ms16ms
([图片] 4ms)
Swipe1284ms2685ms262ms~250ms
Build for simulatorYes——Yes
Build + Run in one callYes (sequential)——Yes (parallel, ~9s faster)
Structured test resultsPartial——Full xcresult JSON
Failure screenshots from xcresult———Auto-exported
Code coverage per fileBasic——Sorted, filterable
Build error diagnosisstderr parsing——xcresult JSON with file:line
Navigate (find + tap + verify)———1 call (~380ms)
Double tap———~60ms
Drag & drop—Coordinates only (3 calls)—Element-to-element (1 call)
Scroll to element—Manual swipe loop—SmartScroll (1 call) [图片]
Alert handling—Single alert—3-tier search + batch accept_all [图片]
iOS 18 ContactsUI dialog———Supported
Batch UI automation———run_plan: multi-step plans with adaptive decisions
Log filteringSubsystem only—PartialTopic-filtered: 90% fewer tokens [图片]
Console log per failed test———Optional
Wait for log pattern———Regex + timeout
Visual regression———Baseline + pixel diff [图片]
Multi-device check———Dark Mode, Landscape, iPad [图片]
Cross-platform (Android)—Yes——
Tools77 (Rust)61 (Node.js + Appium)15 (Swift)58 (Native Swift, 8.5MB)
Cold start~400ms~1000ms~100ms~50ms

Where SilbercueSwift really shines

[killer feat] Screenshots up to 75x faster — ~316ms ([图片] ~15ms)

Free tier screenshots (~316ms) are faster than XcodeBuildMCP (~1127ms) and most alternatives. Pro brings latency down to ~15ms — 5x faster than Appium, 75x faster than XcodeBuildMCP. Agents can take screenshots freely without penalty at either tier.

[killer feat] Structured test results from xcresult bundles — zero guesswork on failures

When a test fails, the agent gets the error message, the exact file:line, a screenshot of the failure state, and optionally the console output — all parsed from Apple's .xcresult format. No guessing from 500 lines of xcodebuild stderr. This is the difference between "agent knows what broke" and "agent guesses what broke".

[killer feat] Single binary, zero dependencies — 58 tools, install in 10 seconds

brew install silbercue/silbercue/silbercueswift — done. 8.5MB native Swift binary. No Node.js, no npm, no Appium server, no Python, no Java, no Rust toolchain. Cold start in ~50ms. The fastest way to get an iOS MCP server running.

[killer feat] Agent reads only what matters — 90% fewer tokens, zero wasted calls (topic filtering [图片])

Free tier already strips noise: 15 known noise processes are excluded at capture time, and duplicate lines are collapsed (79% I/O reduction). Pro adds topic filtering — read_logs categorizes lines into 8 topics and shows only app + crashes by default, with a menu: network(87) lifecycle(12) springboard(8). The agent opens specific topics in one call — no guessing, no iteration.

[strong] One call to dismiss all permission dialogs — 3 alerts in 1 roundtrip [图片]

Every app shows 2–3 permission dialogs on first launch. Other servers require the agent to screenshot → find button → click, per dialog. handle_alert(action: "accept_all") clears them all in a single call, searching across SpringBoard, ContactsUI, and the active app. Free tier handles alerts individually with accept / dismiss.

[strong] Drag & drop with element IDs — 1 call instead of 3 [图片]

"Drag item A above item B" is a single call: drag_and_drop(source_element: "el-0", target_element: "el-1"). The competition only supports raw coordinates, forcing the agent to find both elements, extract their frames, and build a W3C Actions sequence — 3 calls minimum.

[strong] Auto-scroll to off-screen elements — no more manual swipe loops [图片]

find_element(using: "accessibility id", value: "Save", scroll: true) scrolls automatically until the element appears. SmartScroll handles UIKit, SwiftUI, and lazy-loaded lists — no guessing scroll direction.

[strong] View hierarchy in ~31ms (Free) / ~5ms (Pro) — up to 188x faster element inspection

get_source returns the full UI tree in ~31ms (Free) or ~5ms (Pro). The fastest competitor takes 44ms, most take 250ms+. This makes element inspection practically free for agents.

[killer feat] Navigate in one call — find + tap + settle + screenshot in ~380ms

navigate(to: "Settings") finds the element, taps it, waits for the screen to settle, and returns a verification screenshot — all in a single call. No competitor offers this. Agents save 3-4 tool calls per navigation step.

[strong] Batch UI automation — run_plan executes multi-step plans with adaptive decisions

run_plan takes a sequence of UI steps and executes them server-side. When a step needs a decision (unexpected dialog, element not found), it falls back through 4 tiers — from MCP sampling to pause & resume. No more "one tool call per tap" overhead.

Quick Start

Install via Homebrew

bash
brew tap silbercue/silbercuebrew install silbercueswift

Or build from source

bash
git clone https://github.com/silbercue/SilbercueSwift.gitcd SilbercueSwiftswift build -c releasecp .build/release/SilbercueSwift /usr/local/bin/

Configure in Claude Code

One command — installs globally for all projects:

bash
claude mcp add --scope user SilbercueSwift /opt/homebrew/bin/SilbercueSwift

Note: Use the full path (/opt/homebrew/bin/SilbercueSwift). Claude Code starts MCP servers without a full shell PATH, so bare command names won't be found.

Configure in Cursor

Add to ~/.cursor/mcp.json:

json
{  "mcpServers": {    "SilbercueSwift": {      "command": "/opt/homebrew/bin/SilbercueSwift"    }  }}

Configure in other MCP clients (Cline, Continue, etc.)

Any client that supports stdio MCP servers: point to /opt/homebrew/bin/SilbercueSwift with no arguments.

Uninstall

bash
claude mcp remove --scope user SilbercueSwiftbrew uninstall silbercueswiftbrew untap silbercue/silbercue

Free vs Pro

SilbercueSwift ships 49 tools for free — build, test, simulate, automate UI, capture logs, and take screenshots. No time limit, no signup.

Pro adds 9 tools and faster internals for teams and power users who need the full picture.

FreePro
Build, test, sim management49 tools58 tools
Screenshot~316ms~15ms (75x faster)
Structured test results (xcresult)YesYes
Find element31ms<1ms
View hierarchy31ms~5ms
Tap (coordinates)16ms4ms
Click / type / swipe / double tap / long press / drag & dropYesYes
Navigate (find + tap + verify)YesYes
Batch UI automation (run_plan)YesYes
Alert handlingSingle accept/dismiss+ Batch accept_all / dismiss_all
Log captureSmart + verbose+ App mode, topic filtering
Console capture, git toolsYesYes
Scroll to element—SmartScroll
Visual regression—Baseline + pixel diff
Multi-device check—Dark Mode, Landscape, iPad
Accessibility check—Dynamic Type rendering
Localization check—Multi-language + RTL
Pinch / zoom—Yes

Pro costs 12 EUR/month. Get a license on Polar.sh, then:

bash
silbercueswift activate <YOUR-LICENSE-KEY>

58 Tools in 14 Categories

Build (5 tools)

ToolDescription
build_simBuild for iOS Simulator — returns structured errors + caches bundle ID & app path
build_run_simBuild + boot + install + launch in one call — parallel 2-phase pipeline, ~9s faster than sequential
cleanClean build artifacts
discover_projectsFind .xcodeproj/.xcworkspace files
list_schemesList available schemes

Testing & Diagnostics (4 tools)

ToolDescription
test_simRun tests + structured xcresult summary (pass/fail/duration)
test_failuresFailed tests with error messages, file:line, and failure screenshots
test_coverageCode coverage per file, sorted and filterable
build_and_diagnoseBuild + structured errors/warnings from xcresult

Simulator (12 tools)

ToolDescription
list_simsList available simulators
boot_simBoot a simulator
shutdown_simShut down a simulator
install_appInstall .app bundle
launch_appLaunch app by bundle ID
terminate_appTerminate running app
clone_simClone an existing simulator
erase_simErase simulator content and settings
delete_simDelete a simulator
set_orientationRotate device (PORTRAIT, LANDSCAPE_LEFT, LANDSCAPE_RIGHT) via WDA
sim_statusSimulator state (booted/shutdown, device type, runtime)
sim_inspectDetailed simulator info (data path, log path, UDID)

UI Automation (16 tools)

Native input for gestures, WDA for element queries and alerts — no Appium, no Node.js, no Python.

ToolDescriptionLatency
handle_alertAccept, dismiss, or batch-handle system & in-app alerts~200ms
find_element / find_elementsFind elements by accessibility ID, predicate, class chain. scroll: true auto-scrolls until the element appears (SmartScroll — 3 fallback strategies)31ms
([图片] <1ms)
click_elementTap a UI element~75ms
tap_coordinatesCoordinate-based tap~16ms
([图片] ~4ms)
double_tap / long_pressDouble tap or long press at coordinates~60ms / ~1000ms
swipeDirectional swipe~250ms
pinchZoom in/out~400ms [图片]
drag_and_dropDrag from source to target — element-to-element, coordinates, or mixed. Smart defaults for reorderable lists, Kanban boards, sliders~1.3s
navigateFind + tap + settle + screenshot in 1 call — saves 3-4 roundtrips~380ms
type_text / get_textType into or read from elements~100-300ms
get_sourceFull view hierarchy (JSON/XML)~31ms
([图片] ~5ms)
wda_status / wda_create_sessionWDA health check & session management~50-100ms
handle_alert — the smartest alert handler
bash
# Accept a single alert with smart defaultshandle_alert(action: "accept")
# Dismiss with a specific button labelhandle_alert(action: "dismiss", button_label: "Not Now")
# Batch-accept ALL alerts after app launch (unique to SilbercueSwift)handle_alert(action: "accept_all")

3-tier alert search — finds alerts across:

  1. Springboard — system permission dialogs (Location, Camera, Tracking)
  2. ContactsUI — iOS 18+ Contacts "Limited Access" dialog (separate process)
  3. Active app — in-app UIAlertController dialogs

Smart defaults — knows which button to tap:

  • Accept: "Allow" → "Allow While Using App" → "OK" → "Continue" → last button
  • Dismiss: "Don't Allow" (handles Unicode U+2019) → "Cancel" → "Not Now" → first button

Batch mode — accept_all / dismiss_all loops through multiple sequential alerts server-side. One HTTP roundtrip instead of N. Returns details of every handled alert.

These capabilities go beyond what other iOS MCP servers currently offer.

Screenshots (1 tool)

ToolLatency
screenshotFree: ~316ms / Pro: ~15ms

Logs (4 tools)

ToolDescription
start_log_captureSmart-filtered os_log stream — 3 modes: smart (default, topic filtering enabled), app (tight stream, auto-detected), verbose (unfiltered). Deduplicates repetitive lines.
stop_log_captureStop capture
read_logsTopic-filtered reading — default: app + crashes only. Response includes topic menu with line counts. Add topics via include parameter.
wait_for_logWait for regex pattern with timeout — eliminates sleep() hacks
Smart Log Filtering — 4 layers, zero config
bash
# Start capture (default: smart mode — broad stream, topic filtering enabled)start_log_capture()
# Read logs — default shows only app logs + crashes + topic menuread_logs()# → --- 230 buffered, 42 shown [app, crashes] ---# → Topics: app(35) crashes(2) | network(87) lifecycle(12) springboard(8) widgets(0) background(3) system(83)# → Hint: include=["network"] to add SSL/TLS + background transfer logs# → ---# → [42 filtered lines]
# Agent sees network(87) and wants SSL details — one call:read_logs(include: ["network"])
# Narrow stream for production monitoring:start_log_capture(mode: "app")
# Bypass mode logic with explicit predicate:start_log_capture(subsystem: "com.apple.SwiftUI")

4 filter layers:

  1. Stream-side noise exclusion — 15 known noise processes + subsystem/category exclusions removed before buffering. Server-side filtering in logd — 79% I/O reduction.
  2. 3 capture modes — smart (default, broad stream for topic filtering), app (tight, auto-detected bundle ID + process name), verbose (unfiltered).
  3. Read-time topic filtering — read_logs categorizes every buffered line into 8 topics (app, crashes, network, lifecycle, springboard, widgets, background, system). Default shows only app + crashes. Agent adds topics as needed — stateless per call.
  4. Buffer deduplication — 60 identical heartbeat lines become 2: the line itself + ... repeated 59x.

8 topics with LLM-optimized menu:

TopicMatchesUse case
app (always on)subsystem == bundleId OR process == appNameYour app: os_log, print(), NSLog()
crashes (always on)fault-level logsCrashes from any process
networktrustd, nsurlsessiondSSL/TLS certs, background transfers
lifecyclerunningboardd, com.apple.runningboard.*Jetsam, memory pressure, app kills
springboardSpringBoardPush notifications, app state
widgetschronodWidgetKit timeline, refresh budget
backgroundcom.apple.xpc.activity.*BGTaskScheduler, background fetch
systemeverything elseWARNING: high volume

Console (3 tools)

ToolDescription
launch_app_consoleLaunch app with stdout/stderr capture
read_app_consoleRead console output
stop_app_consoleStop console capture

Git (5 tools)

ToolDescription
git_status / git_diff / git_logRead operations
git_commit / git_branchWrite operations

Visual Regression (2 tools)

ToolDescription
save_visual_baselineSave a screenshot as a named baseline
compare_visualCompare current screen against baseline — pixel diff + match score

Multi-Device (1 tool)

ToolDescription
multi_device_checkRun visual checks across multiple simulators (Dark Mode, Landscape, iPad) — returns layout scores

Accessibility (1 tool) [图片]

ToolDescription
accessibility_checkRender screens across Dynamic Type content size categories — detects truncation and layout issues

Localization (1 tool) [图片]

ToolDescription
localization_checkRender screens across languages including RTL (Arabic, Hebrew) — detects layout breaks

Automation (2 tools)

ToolDescription
run_planExecute a multi-step UI automation plan server-side — adaptive decisions with 4-tier fallback
run_plan_decideResume a paused plan with a decision — for clients without MCP sampling

Session (1 tool)

ToolDescription
set_defaultsSet default project, scheme, simulator — avoids repeating params

xcresult Parsing — The Killer Feature

The Problem

Every Xcode MCP server returns raw xcodebuild output. For a test run, that's 500+ lines of noise. AI agents can't reliably extract which tests failed and why.

The Solution

SilbercueSwift uses xcresulttool to parse the .xcresult bundle — the same structured data Xcode's Test Navigator uses.

# One call, structured resulttest_sim(project: "MyApp.xcodeproj", scheme: "MyApp")
→ Tests FAILED in 15.2s  12 total, 10 passed, 2 FAILED  FAIL: Login shows error message    LoginTests.swift:47: XCTAssertTrue failed  FAIL: Profile image loads    ProfileTests.swift:112: Expected non-nil value
  Failure screenshots (2):    /tmp/ss-attachments/LoginTests_failure.png    /tmp/ss-attachments/ProfileTests_failure.png
  Device: iPhone 16 Pro (18.2)  xcresult: /tmp/ss-test-1774607917.xcresult

The agent gets:

  • Pass/fail counts — immediate overview
  • Failure messages with file:line — actionable
  • Failure screenshots — visual context (Claude is multimodal)
  • xcresult path — reusable for test_failures or test_coverage

Deep Failure Analysis

test_failures(xcresult_path: "/tmp/ss-test-*.xcresult", include_console: true)
→ FAIL: Login shows error message [LoginTests/testErrorMessage()]    LoginTests.swift:47: XCTAssertTrue failed    Screenshot: /tmp/ss-attachments/LoginTests_failure.png    Console:      [LoginService] Network timeout after 5.0s      [LoginService] Retrying with fallback URL...      ✘ Test "Login shows error message" failed after 6.2s

Code Coverage

test_coverage(project: "MyApp.xcodeproj", scheme: "MyApp", min_coverage: 80)
→ Overall coverage: 72.3%
  Target: MyApp.app (74.1%)      0.0% AnalyticsService.swift     45.2% LoginViewModel.swift     67.8% ProfileManager.swift
  Target: MyAppTests.xctest (62.0%)     ...

Benchmarks

Measured on M3 MacBook Pro, iOS 26.4 Simulator. All values are median of 5 runs after 2 warmups.

ActioniosefXcodeBuildMCPAppium-MCPSS FreeSS Pro
Screenshot83ms1127ms77ms316ms15ms
Find element50msN/A76ms31ms<1ms
Tap (coordinates)48ms235ms470ms16ms4ms
Swipe262ms1284ms2685ms~250ms~250ms
View hierarchy44ms259ms938ms31ms5ms
Navigate (1 call)———~380ms~380ms
Double tap———~84ms~60ms
Drag & drop—coords only—~1.3s~1.3s
Handle alert——118ms~200ms~200ms
Handle 3 alerts (batch)——3 calls~800ms (1 call)~800ms (1 call)
Scroll to element——swipe loop—Automatic
Build (clean)—2501ms—3188ms1800ms
Simulator list12ms567ms—15ms15ms
Cold start~100ms~400ms~1000ms~50ms~50ms
Binary size~5MB~4MB~200MB8.5MB8.5MB

Comparison with other MCP servers

See feature comparison table above for a detailed breakdown vs XcodeBuildMCP, Appium-MCP, and iosef. All three are excellent projects that pioneered iOS MCP tooling. SilbercueSwift combines all their feature sets with deeper integration into a single native binary. The only trade-off: SilbercueSwift is iOS-only (no Android, watchOS, tvOS, or visionOS).

Architecture

SilbercueSwift (8.5MB Swift binary)├── MCP SDK (modelcontextprotocol/swift-sdk)├── StdioTransport (JSON-RPC)└── 58 Tools in 14 Categories    Build · Test · Simulator · Screenshot · UI Automation    Logs · Console · Visual Regression · Multi-Device    Accessibility · Localization · Automation · Git · Session

No Node.js. No Python. No Appium server. No Selenium. One binary.

Requirements

  • macOS 13+
  • Xcode 15+ (for xcresulttool and simctl)
  • Swift 6.0+ (for building from source)
  • WebDriverAgent installed on simulator (for UI automation tools)

License

The core binary and all 49 free tools are MIT licensed — see LICENSE. Use them however you want, commercially or otherwise.

Pro tools (9 additional tools + faster internals) require a paid license. The license validation code (LicenseManager.swift) is included in the source for transparency — you can see exactly what it checks and when.

Contributing

Issues and pull requests welcome. See CONTRIBUTING.md for guidelines.

Privacy

SilbercueSwift runs entirely on your Mac and collects no user data. See PRIVACY.md for details.


Automating a browser too? Public Browser is the same idea for Chrome.

来源:README.md,提交 b6ceba4

工具

0
SourceWeft 尚未在沙箱中启动过该软件包,因此其工具未知。

版本历史

1
  1. v3.6.3最新Sep 16, 2026

评分

已满足适用检查项的 50%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手;SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分