测试生成

Name: 测试生成
Author: CharlesWiltgen

工程开发已审计

作者仓库星标 977
叉子 74
许可证 MIT
作者更新于 2026年6月15日 03:09
作者仓库 Axiom

流狐档案事实层信息：领域、Agent、信任分、运行环境、权限与网络

领域: 工程开发
兼容 Agent: Claude Code
Cursor
Cline
Codex
Windsurf
Gemini CLI

+20
信任分: 94 / 100 · 已通过审计
作者 / 版本 / 许可: @CharlesWiltgen · MIT
Token 消耗评级: 较高消耗
接入复杂程度: 需简单配置
是否需要外部 API Key: 不需要
兼容的系统: 未声明（默认跨平台）
底层运行要求: 无特殊要求
文件与系统权限: 只读
允许写入 / 修改
网络行为: 仅限本地
安装命令数: 26 条

档案由构建时根据 SKILL.md 与安装命令自动衍生，可能与作者实际意图存在差异。

需要注意： 未限定 allowed-tools，默认拥有全部工具权限。

输出预览 axiom-build.preview

---
name: axiom-build
description: Use when ANY iOS build fails, test crashes, Xcode misbehaves, or environment issue occurs before…
category: 工程开发
runtime: 无特殊运行时
---

# axiom-build 输出预览

## PART A: 任务判断
- 适用问题：代码实现、重构、调试或代码审查。
- 输入要求：目标材料、限制条件、期望输出和验收方式。
- 证据边界：围绕“When to Use / Routing Logic / 1. Environment-First Issues → xcode-debugging”读取原文规则，不把推断写成作者承诺。

## PART B: 执行结果
- **01** 任务判断：确认你的需求是否属于代码实现、重构、调试或代码审查，并标出输入、限制和预期结果。
- **02** 执行计划：优先按“When to Use / Routing Logic / 1. Environment-First Issues → xcode-debugging”拆成步骤，说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果：给出可复制的命令、文件改动、检查清单或内容草稿，并说明如何继续迭代。
- **04** 风险边界：结合 读取文件、写入/修改文件、主要在本地完成、通常不需要额外 API Key 给出执行前确认项。

## Running Rules
- 读取文件、写入/修改文件；主要在本地完成；通常不需要额外 API Key。
- 先小样例验证，再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。

SKILL.md SKILL.md

---
name: axiom-build
description: Use when ANY iOS build fails, test crashes, Xcode misbehaves, or environment issue occurs before…
category: 工程开发
source: CharlesWiltgen/Axiom
---

# axiom-build

## 什么时候使用
- 构建失败先排查环境，避免误把 Xcode 问题当代码缺陷 适合处理工程开发场景下的代码实现、调试、重构、测试或代码审查，核心价值是把输入、判断、执行、验证和交付边界固定下来，避免 Agent 泛泛回答。 把任务拆成可执行、可检查、可继续…
- 面向代码实现、重构、调试或代码审查，优先处理能明确输入、步骤和验收标准的工作。

## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。

## 执行规则
- 围绕「When to Use / Routing Logic / 1. Environment-First Issues → xcode-debugging」组织步骤，不把推断写成作者事实。
- 读取文件、写入/修改文件；主要在本地完成；通常不需要额外 API Key。
- 先跑小样例，确认结果可检查后再扩大任务范围。

## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown，不编造命令、平台或依赖。

证据边界与执行链路

作者原文负责流程事实；仓库文件负责来源和命令；流狐只补充适用场景、限制和质量判断。

skill "axiom-build" {
  输入层     -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
  上下文层   -> When to Use / Routing Logic / 1. Environment-First Issues → xcode-debugging
  规则层     -> SKILL.md 触发条件 / 执行顺序 / 输出格式
  运行层     -> 无特殊运行时 | 读取文件、写入/修改文件 | 主要在本地完成
  安全层     -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
  输出层     -> 可复制结果 + 检查清单 + 下一步迭代
}

解读按原文结构重写，命令、链接、术语均保留；右侧可核对作者原始 SKILL.md

先判断是否适合

核心用途：构建失败先排查环境，避免误把 Xcode 问题当代码缺陷适合处理工程开发场景下的代码实现、调试、重构、测试或代码审查，核心价值是把输入、判断、执行、验证和交付边界固定下来，避免 Agent 泛泛回答。把任务拆成可执行、可检查、可继续迭代的步骤；通常不需…
适用任务：更适合工程开发场景下的代码实现、调试、重构、测试或代码审查，尤其是输入、步骤、验收口径可以反复复用的场景。
不宜硬套：如果任务只有一次性问答、来源材料很少，或需要作者没有说明的命令和外部服务，应先降级为普通对话。

作者设计意图

触发依据：使用时机、上下文收集、输出格式和失败处理共同决定它是否适合当前任务。
来源边界：作者原文负责事实，仓库文件负责命令和文件证据；Fluxly 只补充用途、限制和使用判断。

作者的方法与取舍

执行主线：任务围绕「When to Use / Routing Logic / 1. Environment-First Issues → xcode-debugging」组织，把触发条件、必要上下文、执行规则和交付标准连成一条可复核的链路。
设计重点：重点不在重复教程，而在解释作者为什么规定这些步骤、哪些条件不可省略，以及结果偏离时应回到哪一层检查。
使用判断：先确认任务与原文边界一致，再决定是否启用；来源没有说明的命令、平台和依赖保持未知，不由 Fluxly 补写。

边界和复核

依赖条件：通常不需要额外 API Key，适合先用小任务验证触发是否稳定。
权限提醒：会涉及 read / write 权限，执行前要说明可读写范围、命令边界和回退方式。
质量标准：结果应包含明确产物、必要证据和下一步动作；如果只是泛泛解释，需要补充输入和验收条件后重跑。

Build & Environment

You MUST use this skill for ANY build, environment, or Xcode-related issue before debugging application code.

When to Use

Use this router when you encounter:

Build failures (BUILD FAILED, compilation errors, linker errors)
Test crashes or hangs
Simulator issues (won't boot, device errors)
Xcode misbehavior (stale builds, zombie processes)
Dependency conflicts (CocoaPods, SPM)
Build performance issues (slow compilation)
Environment issues before debugging code

Routing Logic

This router invokes specialized skills based on the specific issue:

1. Environment-First Issues → xcode-debugging

Triggers:

BUILD FAILED without obvious code cause
Tests crash in clean project
Simulator hangs or won't boot
"No such module" after SPM changes
Zombie xcodebuild processes
Stale builds (old code still running)
Clean build differs from incremental build
Device Hub / predicted-vs-built issues in Xcode 27 (OS27)
Reproducing a device-only bug on a simulator (Device Hub) (OS27)

Why xcode-debugging first: 90% of mysterious issues are environment, not code. Check this BEFORE debugging code.

Invoke: skills/xcode-debugging.md

2. Slow Builds → build-performance

Triggers:

Compilation takes too long
Type checking bottlenecks
Want to optimize build time
Build Timeline shows slow phases

Invoke: skills/build-performance.md

3. SPM Dependency Conflicts → spm-conflict-resolver (Agent)

Triggers:

SPM resolution failures
"No such module" after adding package
Duplicate symbol linker errors
Version conflicts between packages
Swift 6 package compatibility issues
Package.swift / Package.resolved conflicts

Why spm-conflict-resolver: Specialized agent that analyzes Package.swift and Package.resolved to diagnose and resolve Swift Package Manager conflicts.

Invoke: Launch spm-conflict-resolver agent

4. Security & Privacy Audit → security-privacy-scanner (Agent)

Triggers:

App Store submission prep
Privacy Manifest requirements (iOS 17+)
Hardcoded credentials in code
Sensitive data storage concerns
ATS violations
Required Reason API declarations

Why security-privacy-scanner: Specialized agent that scans for security vulnerabilities and privacy compliance issues.

Invoke: Launch security-privacy-scanner agent or /axiom:audit security

5. iOS 17→18 Modernization → modernization-helper (Agent)

Triggers:

Migrate ObservableObject to @Observable
Update @StateObject to @State
Adopt modern SwiftUI patterns
Deprecated API cleanup
iOS 17+ migration

Why modernization-helper: Specialized agent that scans for legacy patterns and provides migration paths with code examples.

Invoke: Launch modernization-helper agent or /axiom:audit modernization

6. Build Failure Auto-Fix → build-fixer (Agent)

Triggers:

BUILD FAILED with no clear error details
Build sometimes succeeds, sometimes fails
App builds but runs old code
"Unable to boot simulator" error
Want automated environment-first diagnostics

Why build-fixer: Autonomous agent that checks zombie processes, Derived Data, SPM cache, and simulator state before investigating code. Saves 30+ minutes on environment issues.

Invoke: Launch build-fixer agent or /axiom:fix-build

7. Slow Build Optimization → build-optimizer (Agent)

Triggers:

Builds take too long
Want to identify slow type checking
Expensive build phase scripts
Suboptimal build settings
Want parallelization opportunities

Why build-optimizer: Scans Xcode projects for build performance optimizations — slow type checking, expensive scripts, suboptimal settings — to reduce build times by 30-50%.

Invoke: Launch build-optimizer agent or /axiom:optimize-build

8. General Dependency Issues → build-debugging

Triggers:

CocoaPods resolution failures
"Multiple commands produce" errors
Framework version mismatches
Non-SPM dependency graph conflicts

Invoke: skills/build-debugging.md

9. TestFlight Crash Triage → testflight-triage

Triggers:

Beta tester reported a crash
Crash reports in Xcode Organizer
Crash logs aren't symbolicated
TestFlight feedback with screenshots
App was killed but no crash report

Why testflight-triage: Systematic workflow for investigating TestFlight crashes and reviewing beta feedback. Covers symbolication, crash interpretation, common patterns, and Claude-assisted analysis.

Invoke: See axiom-shipping (skills/testflight-triage.md)

Triggers:

How to find crashes in App Store Connect
ASC metrics dashboard navigation
Understanding crash-free users percentage
Comparing crash rates between versions
Exporting crash data from ASC
App Store Connect API for crash data

Why app-store-connect-ref: Reference for navigating ASC crash analysis, metrics dashboards, and data export workflows.

Invoke: See axiom-shipping (skills/app-store-connect-ref.md)

11. Crash Log Analysis → crash-analyzer (Agent)

Triggers:

User has .ips or .crash file to analyze
User pasted crash report text
Need to parse crash log programmatically
Identify crash pattern from exception type
Check symbolication status

Why crash-analyzer: Autonomous agent that parses crash reports, identifies patterns (null pointer, Swift runtime, watchdog, jetsam), and generates actionable analysis.

Invoke: Launch crash-analyzer agent or /axiom:analyze-crash

12. MetricKit API Reference → metrickit-ref

Triggers:

MetricKit setup and subscription
MXMetricPayload parsing (CPU, memory, launches, hitches)
MXDiagnosticPayload parsing (crashes, hangs, disk writes)
MXCallStackTree decoding and symbolication
Field crash/hang collection
Background exit metrics

Why metrickit-ref: Complete MetricKit API reference with setup patterns, payload parsing, and integration with crash reporting systems.

Invoke: See axiom-performance (skills/metrickit-ref.md)

13. Hang Diagnostics → hang-diagnostics

Triggers:

App hangs or freezes
Main thread blocked for >1 second
UI unresponsive to touches
Xcode Organizer shows hang diagnostics
MXHangDiagnostic from MetricKit
Watchdog terminations (app killed during launch/background transition)

Why hang-diagnostics: Systematic diagnosis of hangs with decision tree for busy vs blocked main thread, tool selection (Time Profiler, System Trace), and 8 common hang patterns with fixes.

Invoke: See axiom-performance (skills/hang-diagnostics.md)

14. Live Debugging → lldb

Triggers:

Need to reproduce a crash interactively
Want to set breakpoints and inspect state
Crash report analyzed, now need live investigation
Need to attach debugger to running app

Why lldb: Crash reports tell you WHAT crashed. LLDB tells you WHY.

Invoke: skills/lldb.md

16. Runtime Console Capture → xclog-ref

Triggers:

Need to see what the app is logging at runtime
App crashes but no crash report (need console output)
Silent failures (network, data, auth) with no UI feedback
Want to capture print()/os_log() output from simulator
Need structured log output for analysis
"What is the app printing?"

Why xclog-ref: Xcode's debug console isn't accessible externally. xclog combines simctl stdout/stderr with log stream JSON to capture everything print(), NSLog(), os_log(), and Logger emit — with structured fields (level, subsystem, category) for automated analysis.

Invoke: /axiom:console

15. Code Signing Issues → code-signing

Triggers:

"No signing certificate found"
"Provisioning profile doesn't include signing certificate"
errSecInternalComponent in CI
ITMS-90035 Invalid Signature on upload
Ambiguous identity / multiple certificates
Entitlement mismatch or missing capability
Setting up CI/CD code signing (GitHub Actions, fastlane match)
Certificate expired or revoked

Why code-signing: Code signing errors are NEVER code bugs — they are 100% configuration (certificates, profiles, entitlements, keychains). Diagnosing with CLI tools takes 5 minutes vs hours of guessing.

Invoke: See axiom-security (skills/code-signing.md) (workflows) or See axiom-security (skills/code-signing-diag.md) (troubleshooting)

Decision Tree

Mysterious/intermittent/clean build fails? → xcode-debugging (environment-first)
SPM dependency conflict? → spm-conflict-resolver (Agent)
CocoaPods/other dependency conflict? → build-debugging
Slow build time? → build-performance
Security/privacy/App Store prep? → security-privacy-scanner (Agent)
Want automated build fix (environment-first diagnostics)? → build-fixer (Agent)
Want build time optimization scan? → build-optimizer (Agent)
Modernization/deprecated APIs? → modernization-helper (Agent)
TestFlight crash/feedback? → testflight-triage
Navigating App Store Connect? → app-store-connect-ref
Have a crash log (.ips/.crash)? → crash-analyzer (Agent)
MetricKit setup/parsing? → metrickit-ref
App hang/freeze/watchdog? → hang-diagnostics
Need to reproduce crash interactively / inspect runtime state? → lldb
Code signing error (certificate, profile, entitlement, Keychain)? → code-signing / code-signing-diag
Need to see runtime console output (print/os_log)? → xclog-ref or /axiom:console

Anti-Rationalization

Thought	Reality
"I know how to fix this linker error"	Linker errors have 4+ root causes. xcode-debugging diagnoses all in 2 min.
"Let me just clean the build folder"	Clean builds mask the real issue. xcode-debugging finds the root cause.
"It's just an SPM issue, I'll fix Package.swift"	SPM conflicts cascade. spm-conflict-resolver analyzes the full dependency graph.
"The simulator is just slow today"	Simulator issues indicate environment corruption. xcode-debugging checks systematically.
"I'll skip environment checks, it compiles locally"	Environment-first saves 30+ min. Every time.
"I'll read the crash report more carefully instead of reproducing"	Crash reports show WHAT crashed, not WHY. Reproducing in LLDB with breakpoints reveals the actual state. `skills/lldb.md` has the workflow.
"I know my certificate is fine, let me check the code"	Code signing errors are NEVER code bugs. 100% configuration. code-signing diagnoses with CLI in 5 min.
"I can't see what the app is logging without Xcode"	xclog captures print() + os_log from the simulator. Structured JSON output with level, subsystem, category. `/axiom:console`.

When NOT to Use (Conflict Resolution)

Do NOT use axiom-build for these — use the correct router instead:

Error Type	Correct Router	Why NOT axiom-build
Swift 6 concurrency errors	`/skill axiom-concurrency`	Code error, not environment
SwiftData migration errors	`/skill axiom-data`	Schema issue, not build environment
"Sending 'self' risks data race"	`/skill axiom-concurrency`	Language error, not Xcode issue
Type mismatch / compilation errors	Fix the code	These are code bugs

axiom-build is for environment mysteries, not code errors:

✅ "No such module" when code is correct
✅ Simulator won't boot
✅ Clean build fails, incremental works
✅ Zombie xcodebuild processes
❌ Swift concurrency warnings/errors
❌ Database migration failures
❌ Type checking errors in valid code

Example Invocations

User: "My build failed with a linker error" → Invoke: skills/xcode-debugging.md (environment-first diagnostic)

User: "Builds are taking 10 minutes" → Invoke: skills/build-performance.md

User: "SPM won't resolve dependencies" → Invoke: spm-conflict-resolver agent

User: "Two packages require different versions of the same dependency" → Invoke: spm-conflict-resolver agent

User: "Duplicate symbol linker error" → Invoke: spm-conflict-resolver agent

User: "I need to prepare for App Store security review" → Invoke: security-privacy-scanner agent

User: "Do I need a Privacy Manifest?" → Invoke: security-privacy-scanner agent

User: "Are there hardcoded credentials in my code?" → Invoke: security-privacy-scanner agent

User: "How do I migrate from ObservableObject to @Observable?" → Invoke: modernization-helper agent

User: "Update my code to use modern SwiftUI patterns" → Invoke: modernization-helper agent

User: "Should I still use @StateObject?" → Invoke: modernization-helper agent

User: "A beta tester said my app crashed" → Invoke: See axiom-shipping (skills/testflight-triage.md)

User: "I see crashes in App Store Connect but don't know how to investigate" → Invoke: See axiom-shipping (skills/testflight-triage.md)

User: "My crash logs aren't symbolicated" → Invoke: See axiom-shipping (skills/testflight-triage.md)

User: "I need to review TestFlight feedback" → Invoke: See axiom-shipping (skills/testflight-triage.md)

User: "How do I find crashes in App Store Connect?" → Invoke: See axiom-shipping (skills/app-store-connect-ref.md)

User: "Where's the crash-free users metric in ASC?" → Invoke: See axiom-shipping (skills/app-store-connect-ref.md)

User: "How do I export crash data from App Store Connect?" → Invoke: See axiom-shipping (skills/app-store-connect-ref.md)

User: "Analyze this crash log" [pastes .ips content] → Invoke: crash-analyzer agent or /axiom:analyze-crash

User: "Parse this .ips file: ~/Library/Logs/DiagnosticReports/MyApp.ips" → Invoke: crash-analyzer agent or /axiom:analyze-crash

User: "Why did my app crash? Here's the report..." → Invoke: crash-analyzer agent or /axiom:analyze-crash

User: "How do I set up MetricKit to collect crash data?" → Invoke: See axiom-performance (skills/metrickit-ref.md)

User: "How do I parse MXDiagnosticPayload?" → Invoke: See axiom-performance (skills/metrickit-ref.md)

User: "What's in MXCallStackTree and how do I decode it?" → Invoke: See axiom-performance (skills/metrickit-ref.md)

User: "My app hangs sometimes" → Invoke: See axiom-performance (skills/hang-diagnostics.md)

User: "The main thread is blocked and UI is unresponsive" → Invoke: See axiom-performance (skills/hang-diagnostics.md)

User: "Xcode Organizer shows hang diagnostics for my app" → Invoke: See axiom-performance (skills/hang-diagnostics.md)

User: "My app was killed by watchdog during launch" → Invoke: See axiom-performance (skills/hang-diagnostics.md)

User: "I have a crash report and need to reproduce it in the debugger" → Invoke: skills/lldb.md

User: "How do I set breakpoints to catch this crash?" → Invoke: skills/lldb.md

User: "My build is failing with BUILD FAILED but no error details" → Invoke: build-fixer agent or /axiom:fix-build

User: "Build sometimes succeeds, sometimes fails" → Invoke: build-fixer agent or /axiom:fix-build

User: "How can I speed up my Xcode build times?" → Invoke: build-optimizer agent or /axiom:optimize-build

User: "No signing certificate found when I try to build" → Invoke: See axiom-security (skills/code-signing-diag.md)

User: "errSecInternalComponent in my GitHub Actions CI" → Invoke: See axiom-security (skills/code-signing-diag.md)

User: "How do I set up code signing for GitHub Actions?" → Invoke: See axiom-security (skills/code-signing.md)

User: "What is my app printing to the console?" → Invoke: /axiom:console

User: "I need to see the simulator console output" → Invoke: /axiom:console

User: "The app fails silently, no error in the UI" → Invoke: /axiom:console

安装命令

一键复制安装选择包管理器,点击右侧复制,粘贴到终端回车即可

npx skills add CharlesWiltgen/Axiom/axiom-codex/skills/axiom-build -y

查看通用安装教程

查看 Skill 原文

兼容工具

Claude Code
Cursor
Cline
Codex
Windsurf
Gemini
OpenCode
Roo Code
AAide
+Augment
GitHub Copilot
OpenClaw
KKiro
Junie
WWarp
Goose
Trae
Kilo Code
OpenHands
Amp
Zencoder
CContinue
AIAider
CodeBuddy
CodeGeeX
Replit Agent

安全审计

94 / 100

已审计

未声明 version。
未限定 allowed-tools，默认拥有全部工具权限。

元信息

来源: local
CharlesWiltgen/Axiom
许可证: MIT
English

@CharlesWiltgen 的其他技能

分类与标签

分类工程开发

测试生成

先判断是否适合

作者设计意图

作者的方法与取舍

边界和复核

Build & Environment

When to Use

Routing Logic

1. Environment-First Issues → xcode-debugging

2. Slow Builds → build-performance

3. SPM Dependency Conflicts → spm-conflict-resolver (Agent)

4. Security & Privacy Audit → security-privacy-scanner (Agent)

5. iOS 17→18 Modernization → modernization-helper (Agent)

6. Build Failure Auto-Fix → build-fixer (Agent)

7. Slow Build Optimization → build-optimizer (Agent)

8. General Dependency Issues → build-debugging

9. TestFlight Crash Triage → testflight-triage

10. App Store Connect Navigation → app-store-connect-ref

11. Crash Log Analysis → crash-analyzer (Agent)

12. MetricKit API Reference → metrickit-ref

13. Hang Diagnostics → hang-diagnostics

14. Live Debugging → lldb

16. Runtime Console Capture → xclog-ref

15. Code Signing Issues → code-signing

Decision Tree

Anti-Rationalization

When NOT to Use (Conflict Resolution)

Example Invocations

讨论