qa-test-skills
从需求文档自动生成结构化测试用例,覆盖功能测试、边界分析、组合测试和回归测试全流程。自动串联48个专家级子技能,按12步工作流编排执行。适用于:上传需求文档(PRD/Word/PDF/URL)需要完整测试用例时、不知道如何设计测试场景或担心遗漏边界条件时、需要AI评审测试输出并补充测试盲区时。每个步骤都有独立技能支撑,输出格式统一、需求可追溯、覆盖率可量化。 触发场景:生成测试用例、帮我测试、设计测试、上传需求、开始测试、获取安装指引时。 Use when the user asks about: generating structured test cases from a PRD, Word, PDF, or URL through an orchestrated workflow covering functional, boundary, combination, and regression testing. Skill: qa-test-skills Owner: kokxi Summary: 从需求文档自动生成结构化测试用例,覆盖功能测试、边界分析、组合测试和回归测试全流程。自动串联48个专家级子技能,按12步工作流编排执行。适用于:上传需求文档(PRD/Word/PDF/URL)需要完整测试用例时、不知道如何设计测试场景或担心遗漏边界条件时、需要AI评审测试输出并补充测试盲区时。每个步骤都有独立技能支撑,输出格式统一、需求可追溯、覆盖率可量化。 触发场景:生成测试用例、帮我测试、设计测试、上传需求、开始测试、获取安装指引时。 Use when the user asks about: generating structured test cases from a PRD, Word, PDF, or URL through an orchestrated workflow covering functional, boundar
Rank
62
Safety
84
Downloads
981
Updated
Oct 11, 2026
Version
1.8.0
Source
CLAWHUB
About
What it does, and when to use it.
Capability contract not published. No trust telemetry is available yet. 981 downloads reported by the source. Last updated 10/11/2026.
Avoid when
- Contract metadata is missing or unavailable for deterministic execution.
Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing
Public facts
Every fact links back to the source it came from.
- Vendor
- Clawhubvendor · observed Oct 11, 2026
- Protocol compatibility
- OpenClawcompatibility · observed Oct 11, 2026
- Adoption signal
- 981 downloadsadoption · observed Oct 11, 2026
- Latest release
- 1.8.0release · observed Sep 29, 2026
- Handshake status
- UNKNOWNsecurity
What changed
Newest first. Only changes that point at a source are listed.
Release 1.8.0release
Version 1.8.0 - Migrated SKILL.md to new metadata structure for improved compatibility and clarity. - Moved key workflow, references, and input/output definitions to metadata fields. - Updated installation and loading instructions for multi-skill orchestration; clarified distinction between core and sub-skills. - Adjusted guidance for platform-specific enhancements and skill file loading. - Removed deprecated skill-card.md file.
Observed Sep 29, 2026 · source
Install and run
Setup complexity: low.
clawhub skill install s170jw3s1atcj5jwhqb4r7v7eh8912kp:qa-test-skills- Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
- Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.
Contract: missing
curl -s "https://www.xpersona.co/api/v1/agents/clawhub-kokxi-qa-test-skills/snapshot"
Run-check
$0.02 USD1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.
Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.
Documentation
CLAWHUB
134,670 characters of source documentation, loaded on request.
Extracted files
5 files captured from the source.
SKILL.md
---
name: qa-test-skills
description: >-
从需求文档自动生成结构化测试用例,覆盖功能测试、边界分析、组合测试和回归测试全流程。自动串联48个专家级子技能,按12步工作流编排执行。适用于:上传需求文档(PRD/Word/PDF/URL)需要完整测试用例时、不知道如何设计测试场景或担心遗漏边界条件时、需要AI评审测试输出并补充测试盲区时。每个步骤都有独立技能支撑,输出格式统一、需求可追溯、覆盖率可量化。 触发场景:生成测试用例、帮我测试、设计测试、上传需求、开始测试、获取安装指引时。 Use when the user asks about: generating structured test cases from a PRD, Word, PDF, or URL through an orchestrated workflow covering functional, boundary, combination, and regression testing.
license: MIT
allowed-tools: Read Grep Glob WebFetch Bash
metadata:
display-name: "QA Test Skills"
version: "1.8.0"
disable-model-invocation: "false"
when-to-use: "用户说\"生成测试用例\"、\"帮我测试\"、\"设计测试\"、\"上传需求\"、\"开始测试\"、上传需求文档/URL时自动激活, 需要完整测试流程时,想浏览技能集目录、了解QA Test Skills包含哪些技能、查看技能分类说明、获取安装指引时"
related-skills: "{\"all_skills\":[\"qa-input-validation\",\"qa-requirement-review\",\"qa-req-deconstruction\",\"qa-risk-intuition\",\"qa-heuristic-checklist\",\"qa-scenario-tree\",\"qa-boundary-deep-dive\",\"qa-combination-strategy\",\"qa-state-transition\",\"qa-domain-modeling\",\"qa-regression-testing\",\"qa-ai-context-engineering\",\"qa-ai-prompt-strategy\",\"qa-ai-output-critique\",\"qa-ai-blindspot-compensation\",\"qa-output-validation\",\"qa-test-reporting\",\"qa-agent-testing\",\"qa-expert-review\",\"qa-api-testing\",\"qa-mobile-testing\",\"qa-specialized-testing\",\"qa-exploratory-testing\",\"qa-tech-debt-management\",\"qa-test-estimation\",\"qa-bug-lifecycle\",\"qa-bug-reporting\",\"qa-bug-root-cause-analysis\",\"qa-execution-observation\",\"qa-ci-cd-testing\",\"qa-code-review-for-test\",\"qa-critical-thinking\",\"qa-question-framework\",\"qa-test-case-design\",\"qa-test-strategy-design\",\"qa-release-risk-governance\",\"qa-quality-metrics\",\"qa-test-automation-arch\",\"qa-tech-selection\",\"qa-testability-advocacy\",\"qa-test-data-engineering\",\"qa-test-env-data\",\"qa-shift-left\",\"qa-shift-right\",\"qa-test-leadership\",\"qa-stakeholder-communication\",\"qa-team-coaching\",\"qa-retrospective\"]}"
references: "[\"references/depth-benchmarks.md\",\"references/enforcement.md\",\"references/format.md\",\"references/routing.md\",\"references/workflow-detail.md\"]"
input-format: "{\"required\":[{\"name\":\"用户需求\",\"type\":\"string\",\"description\":\"用户的需求描述,可以是文字、文件路径或URL\"}],\"optional\":[{\"name\":\"附件\",\"type\":\"file\",\"description\":\"上传的需求文档\"},{\"name\":\"URL\",\"type\":\"string\",\"description\":\"需求文档链接\"}]}"
output-format: "{\"structure\":[{\"test_cases\":\"测试用例列表\"},{\"coverage_report\":\"覆盖率报告\"},{\"risk_areas\":\"风险区域\"},{\"test_report\":\"测试报告\"}],\"traceability\":[\"每个测试用例带唯一ID(TC_{模块缩写}_{功能缩写}_{序号},如 TC_API_LOGIN_001)\",\"关联需求ID(REQ-XXXX)\",\"关联场景ID(SC-XXXX)\"]}"
---
# QA Test Skills — 测试工作流编排引擎
> ⚠️ **本技能集共 49 个技能(本入口 + 48 个子技能)**。
> 本入口技能安装后,请补充安装其余 48 个子技能以获得完整 12 步工作流体验:
>
> ```bash
> # 推荐:一次安装全套(skills.sh)
> npx skills add Kokxi/qa-test-skills
>
> # 或手动:复制 skills/ 目录下 49 个子目录到 Agent 的 skills 目录
> cp -r qa-test-skills/skills/* ~/.claude/skills/ # Claude Code
> _meta.json
{
"ownerId": "kn71y9b23csfx0ykgm55d5m9x5891zt8",
"slug": "qa-test-skills",
"version": "1.8.0",
"publishedAt": 1790655877064
}references/depth-benchmarks.md
# 深度量化基准来源说明 > 本文档说明各 skill 元数据中 `metadata.depth-requirement` 的 reference_value 乘数基准(简单×N / 中等×N / 复杂×N)来源,让 AI 执行时能判断该按哪个乘数生成,也让新人理解乘数背后的逻辑。 ## 一、为什么需要深度量化基准 AI 执行测试技能时面临一个两难: - **乘数过低** → 输出偷懒(如风险评估只识别 2 个风险点就收工) - **乘数过高** → 过度设计(如简单 CRUD 功能强制生成 7 层边界分析) `reference_value` 的乘数就是给 AI 的"档位标尺":根据需求复杂度选择对应深度档位,避免输出过浅或过深。 ## 二、乘数档位的来源 这些乘数不是凭空拍的,而是基于三个来源的交叉校准: ### 来源 1:测试行业经验基准 来自一线测试团队的历史数据回溯(约 200+ 项目迭代): | 复杂度档位 | 典型特征 | 测试点密度(个/需求点) | |-----------|---------|----------------------| | 简单 | 单表 CRUD、静态展示、无外部依赖 | 1-2 | | 中等 | 多表关联、有业务规则、1-2 个集成点 | 2-3 | | 复杂 | 状态机、并发、分布式、资金/安全链路 | 3-4+ | **对应到 reference_value**: - `简单×1 / 中等×2 / 复杂×3` ← 基础产出型 skill(如 bug-reporting、ci-cd-testing) - `简单×2 / 中等×3 / 复杂×4` ← 深度分析型 skill(如 boundary-deep-dive、req-deconstruction) - `简单×3 / 中等×5 / 复杂×7` ← 认知质疑型 skill(如 critical-thinking、question-framework) ### 来源 2:与 `minimum` 字段的倍数关系 每个 skill 同时声明了 `minimum`(最小输出量)和 `reference_value`(档位基准)。两者的倍数关系约束了乘数不能随意调: - `minimum` 通常是"简单档位 ×1"的下限 - "中等档位"至少是"简单档位"的 2 倍 - "复杂档位"至少是"简单档位"的 3 倍 如果某 skill 的 `minimum` 是"至少识别 5 个风险点",那么: - 简单档位 ≈ 5(×1) - 中等档位 ≈ 10(×2) - 复杂档位 ≈ 15(×3) ### 来源 3:同类 skill 的横向对齐 同类 skill 的乘数档位保持一致,避免用户在 skill 之间切换时产生深度落差: | skill 类别 | 典型 skill | 档位模式 | |-----------|-----------|---------| | 认知质疑型 | qa-critical-thinking, qa-question-framework, qa-scenario-tree | ×3 / ×5 / ×7 | | 深度分析型 | qa-boundary-deep-dive, qa-req-deconstruction, qa-test-case-design, qa-domain-modeling, qa-state-transition | ×2 / ×3 / ×4 | | 基础产出型 | 多数专项测试、执行洞察、沟通传承类 | ×1 / ×2 / ×3 | ## 三、乘数不是硬约束,是建议档位 ⚠️ **重要**:`reference_value` 是**建议档位**,不是硬性要求。 - AI 应当根据**实际需求复杂度**判断属于简单/中等/复杂哪一档,再按对应乘数生成 - 如果需求明显是"复杂"档但用户只要求"简单"输出深度,以用户要求为准 - `minimum` 是**硬下限**(必须达到),`reference_value` 是**软建议**(推荐达到) ## 四、复杂度档位判定指引 AI 在执行 skill 前,先判定需求属于哪个复杂度档位: ### 简单档(×1 基准) - 单表 CRUD、静态展示 - 无外部依赖、无并发、无状态流转 - 历史缺陷率低、用户量小 ### 中等档(×2 基准) - 多表关联、有业务规则 - 1-2 个外部集成点 - 有基础的状态流转(如订单状态机) - 中等用户量、偶尔出问题 ### 复杂档(×3 基准) - 资金/安全/核心链路 - 并发、分布式、状态机复杂 - 多个外部依赖、数据契约敏感 - 高用户量、历史缺陷多发 ## 五、基准的可挑战性 这些乘数基准基于 2026 年上半年的项目数据回溯,**不是永恒真理**: - 如果你的团队历史数据表明"简单功能其实需要 ×2 深度",可以调整 - 如果某个 skill 的乘数导致输出过度或不足,欢迎提 issue 反馈 - 基准会随项目迭代和数据积累持续校准 --- **相关文档**: - [`references/enforcement.md`](enforcement.md) — 强制执行要求 - [`references/workflow-detail.md`](workflow-detail.md) — 工作流详细执行指南
references/enforcement.md
# 强制执行要求 **⚠️ 重要:agent必须严格按照以下要求执行,不得跳过任何步骤** --- ## 全局限制 - **默认禁止读取代码**:整个技能集使用期间,默认禁止AI读取代码实现 - **目的**:防止AI偷懒根据代码生成测试用例,确保测试用例验证需求而非代码实现 - **例外情况**:仅在特定技能(如代码评审、Bug根因分析)中明确允许读取代码 --- ## 1. skill加载 ≠ skill执行 - **要求**:加载技能后,必须执行技能要求的实际分析,并产出具体输出 - **禁止**:仅加载技能SKILL.md而不执行分析 - **验证**:每个步骤必须有明确的输出文件或报告 ## 2. 任务创建 ≠ 任务执行 - **要求**:创建任务后,必须执行任务并产出实际内容 - **禁止**:仅创建任务标记完成而不执行 - **验证**:任务完成后必须检查是否有实际产出 ## 3. actor调用失败必须重试 - **要求**:actor调用失败后,必须修复参数并重试 - **禁止**:actor失败后跳过步骤 - **重试策略**:修复参数格式 → 重新调用 → 验证结果 ## 4. 必须按顺序执行所有步骤 - **要求**:严格按照工作流顺序执行,不跳过任何步骤 - **禁止**:跳过步骤7(提示词生成)直接生成用例 - **验证**:每个步骤完成后必须有明确的输出 ## 5. 输出评审是必须步骤 - **要求**:生成用例后,必须执行qa-ai-output-critique和qa-ai-blindspot-compensation - **禁止**:将"生成用例"视为最终目标 - **验证**:评审完成后必须有评审报告和盲区补偿用例
references/format.md
# 标准化输出模板与检查清单
## 1. 表格格式(Markdown)
```markdown
| 用例编号 | 测试类型 | 功能模块 | 测试标题 | 用例级别 | 预置条件 | 测试步骤 | 预期结果 | 风险等级 |
|---------|---------|---------|---------|---------|---------|---------|---------|---------|
| TC_XXX_001 | 功能测试 | 模块名 | 标题描述 | P0 | 条件描述 | 步骤1;步骤2;步骤3 | 结果描述 | 高 |
```
## 2. 用例编号规则
```
格式:TC_{模块缩写}_{功能缩写}_{序号}
示例:
- TC_USER_LOGIN_001 用户模块-登录-第1条
- TC_ORDER_PAY_002 订单模块-支付-第2条
- TC_API_LOGIN_003 接口-登录-第3条
```
## 3. 用例级别定义
| 级别 | 说明 | 占比建议 |
|------|------|---------|
| P0 | 关键:核心业务流程 | ≤20% |
| P1 | 重要:主要功能 | ≤40% |
| P2 | 一般:次要功能 | ≤30% |
| P3 | 可选:边缘场景 | ≤10% |
## 4. 生成后自查
```
用例生成完成后,自动执行自查:
1. 完整性检查
- [ ] 功能点是否覆盖(按现有需求文档口径)?
- [ ] 异常场景是否覆盖?
- [ ] 边界条件是否覆盖?
- [ ] 安全场景是否覆盖?
2. 质量检查
- [ ] 用例标题是否清晰?
- [ ] 预置条件是否完整?
- [ ] 测试步骤是否可执行?
- [ ] 预期结果是否可验证?
3. 规范检查
- [ ] 用例编号是否规范?
- [ ] 用例级别是否合理?
- [ ] 风险等级是否准确?
- [ ] 输出格式是否标准?
4. 平台检查(如有平台专项)
- [ ] 平台特性是否覆盖?
- [ ] 兼容性是否考虑?
- [ ] 性能是否验证?
- [ ] 用户体验是否评估?
```
## 5. 覆盖率要求
> ⚠️ 覆盖率必须标注口径:覆盖的是**现有需求文档**,不得使用"全覆盖""100%"等绝对化表述
> (见入口工作流 SKILL.md 的"覆盖率与缺口诚实性硬约束");缺失模块只能标注"未覆盖+原因"。
| 维度 | 覆盖率要求 |
|------|-----------|
| 功能覆盖率 | 100%需求点(基于现有需求文档) |
| P0用例 | 100%覆盖(P0 核心流程,现有需求范围内) |
| 异常场景 | ≥30%用例 |
| 边界场景 | ≥10%用例 |AionUi
Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!
activepieces
AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents
cherry-studio
AI productivity studio with smart chat, autonomous agents, and 300+ assistants.
CopilotKit
The Frontend for Agents & Generative UI. React + Angular
Machine-readable data
The same record, as JSON, for agents and crawlers.
- Card JSON
- Facts JSON
- Snapshot JSON
- Contract JSON
- Trust JSON
- Changes JSON
- Warnings JSON
- Seen by JSON
- Disagreements JSON
{
"facts": [
{
"factKey": "vendor",
"category": "vendor",
"label": "Vendor",
"value": "Clawhub",
"href": "https://clawhub.ai/kokxi/skills/qa-test-skills",
"sourceUrl": "https://clawhub.ai/kokxi/skills/qa-test-skills",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-11T22:52:02.596Z",
"isPublic": true
},
{
"factKey": "protocols",
"category": "compatibility",
"label": "Protocol compatibility",
"value": "OpenClaw",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-kokxi-qa-test-skills/contract",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-kokxi-qa-test-skills/contract",
"sourceType": "contract",
"confidence": "medium",
"observedAt": "2026-10-11T22:52:02.596Z",
"isPublic": true
},
{
"factKey": "traction",
"category": "adoption",
"label": "Adoption signal",
"value": "981 downloads",
"href": "https://clawhub.ai/kokxi/qa-test-skills",
"sourceUrl": "https://clawhub.ai/kokxi/qa-test-skills",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-11T22:52:02.596Z",
"isPublic": true
},
{
"factKey": "latest_release",
"category": "release",
"label": "Latest release",
"value": "1.8.0",
"href": "https://clawhub.ai/kokxi/qa-test-skills",
"sourceUrl": "https://clawhub.ai/kokxi/qa-test-skills",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-09-29T04:24:37.064Z",
"isPublic": true
},
{
"factKey": "handshake_status",
"category": "security",
"label": "Handshake status",
"value": "UNKNOWN",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-kokxi-qa-test-skills/trust",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-kokxi-qa-test-skills/trust",
"sourceType": "trust",
"confidence": "medium",
"observedAt": null,
"isPublic": true
}
],
"events": [
{
"eventType": "release",
"title": "Release 1.8.0",
"description": "Version 1.8.0 - Migrated SKILL.md to new metadata structure for improved compatibility and clarity. - Moved key workflow, references, and input/output definitions to metadata fields. - Updated installation and loading instructions for multi-skill orchestration; clarified distinction between core and sub-skills. - Adjusted guidance for platform-specific enhancements and skill file loading. - Removed deprecated skill-card.md file.",
"href": "https://clawhub.ai/kokxi/qa-test-skills",
"sourceUrl": "https://clawhub.ai/kokxi/qa-test-skills",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-09-29T04:24:37.064Z",
"isPublic": true
}
]
}Record generated Oct 12, 2026.
For crawlers
This page is free to read. The run-check above is the only paid part, and it answers HTTP 402 until it is paid. Everything else here is public.
- One record, as JSON: card, facts, snapshot, contract, trust, changes, warnings, seen by, disagreements.
- Every agent, one feed: /.well-known/ai-catalog.json
- What this site sells, and the price: /.well-known/x402
- Paid run-check: /api/v1/agents/clawhub-kokxi-qa-test-skills/run-check
