Home / mingdui / ming-qa · skills/quality-assurance-agent/SKILL.md · GitHub

quality-assurance-agent skillA

quality-assurance-agent is agent-read markdown (skill) from mingdui/ming-qa: 端到端 QA 验收编排器。当用户要求做功能验收、回归测试、质量检查、或对某个模块/流程/branch/PR进行系统化测试时触发。 作为顶层路由器,按阶段调度子 skill(上下文收集 → 风险分析 → 用例设计 → 脚本生成 → 执行与修复 → 代码审查 → 报告生成)。 你自己不写测试代码也不直接判断质量——你把每个阶段交给专职子 skill 处理。 用例确认是流程中唯一的强制人工门禁,其余阶段自动衔接。 不适用:单次手动测试、纯开发/代码生成任务、无明确验收范围的技术问答。.

Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.

What the file says

# Quality Assurance Agent — 顶层路由器

> **CLI 调用约定**:本工具包的 CLI 是 `quality-assurance-agent/scripts/qa_agent.py`。
> 它**不以 PATH 命令的形式分发**——命令由你(agent)执行,人不必手敲。
> 开工前解析一次 skill 目录,之后所有命令一律写成
> `python "$QA_AGENT_DIR/scripts/qa_agent.py" <cmd>`:
>
>     QA_AGENT_DIR="${QA_AGENT_CLI:-$(dirname "$(find ~/.claude/skills ~/.agents/skills ~/.codex/skills .claude/skills .agents/skills .codex/skills -maxdepth 2 -name SKILL.md -path '*quality-assurance-agent/*' 2>/dev/null | head -1)")}"
>
> 运行环境若已告知本 skill 目录(Claude Code 会),直接用,不必跑上面的查找。
> 完整命令语法见 `references/cli-reference.md`。

## 你的定位

你不是执行者,是调度者。你的任务是把一个 QA 请求拆成固定阶段,每个阶段交给对应的子 skill 处理。
你自己不写测试、不写用例、不修 bug,不直接判断 Ready——你只负责**按正确顺序、用正确参数调用正确的子 skill**。

整个 QA 链路是管道的、有门禁的、可追溯的。任何阶段发现阻塞问题,记录到对应产物里,不跳过、不替用户做决定。

---

## 调用子 skill 的规则

调用子 skill 使用 `Skill` 工具,skill 名称不带前缀路径。每个子 skill 有完整的独立 SKILL.md,调用时自动加载。

| 阶段 | 子 skill | 职责一句话 |
|---|---|---|
| 0 | `qa-context-profiler` | 收集仓库事实和环境证据,不做任何判断 |
| 1 | `qa-risk-analyzer` | 识别高风险业务路径和必需的验证点 |
| 2 | `qa-testcase-designer` | 生成中文业务用例,等待用户确认(唯一强制人工门禁) |
| 3 | `qa-test-script-generator` | 把已确认用例转为可执行的 spec-task 和测试脚本 |
| 4 | `qa-test-runner` | 执行测试、分类失败、修最小根因、跑 completion 门禁 |
…

Read the whole file at its exact version.

How to install

Latest version
mdr add mingdui/ming-qa/quality-assurance-agent@git:20260920.56e926f
Exact content
mdr add mingdui/ming-qa/quality-assurance-agent@sha256:21e655188d2076cf

Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_jg6y46nw23pmpjgi.svg)](https://markdownregistry.com/a/art_jg6y46nw23pmpjgi)

1 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
git:20260920.56e926f latest2026-09-20 56e926f 20,357 BA view · diff
git:20260918.0165bf32026-09-18 0165bf3 19,962 BA view · diff
git:20260918.a9168aa2026-09-18 a9168aa 18,859 BA view · diff
git:20260918.90b9a152026-09-18 90b9a15 18,316 BA view · diff
git:20260918.8d4fb772026-09-18 8d4fb77 17,580 BA view

Audit of the latest version

A  17 of 17 checks passed. Deterministic, no model, same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (20357 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

mingdui/ming-qa · 28 stars · license MIT · pushed 2026-09-21 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_jg6y46nw23pmpjgi
GET https://markdownregistry.com/api/v1/resolve?ref=mingdui/ming-qa/quality-assurance-agent
GET https://markdownregistry.com/api/v1/blob/21e655188d2076cf2e91ed2b1ff3e71262585f8c420db338d8326f1ef5e6296c

Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

More from mingdui/ming-qa

CLAUDE.md claude
mingdui/ming-qa · CLAUDE.md
git:20260920.56e926f · audit A · 28 stars
qa-code-reviewer skill
mingdui/ming-qa · skills/qa-code-reviewer/SKILL.md · QA 流程倒数第二步的独立审查环节。在 qa-test-runner 执行完成后,以独立视角审查代码本身——检查架构、安全、并发正确性、数据一致性、 回归风险、测试覆盖缺口——不受执行方结论的影响。产出…
git:20260918.2d0e957 · audit A · 28 stars
qa-context-profiler skill
mingdui/ming-qa · skills/qa-context-profiler/SKILL.md · QA 上游阶段——为仓库构建事实画像和环境证据。当你需要收集模块的代码上下文、索引已有用例和测试、 检查前后端服务可达性、准备环境快照时触发。只收集事实,不做风险分析、不写用例、不判断质量。 通常在…
git:20260918.2d0e957 · audit A · 28 stars
qa-report-generator skill
mingdui/ming-qa · skills/qa-report-generator/SKILL.md · QA 流程的收口阶段——汇总执行(completion)、审查(code-review)、就绪(readiness)三个门禁产物,渲染最终 HTML 报告,给出最终就绪判定。 在 qa-test-runner…
git:20260918.2d0e957 · audit A · 28 stars
qa-risk-analyzer skill
mingdui/ming-qa · skills/qa-risk-analyzer/SKILL.md · 在测试用例设计之前,先识别高风险业务路径和必需的验证断言(oracle)。分析需求文档、代码 diff、已有用例和上下文, 找出资金/权限/状态机/并发/异步/数据一致性等高危区域,输出每条风险的…
git:20260920.56e926f · audit A · 28 stars
qa-test-runner skill
mingdui/ming-qa · skills/qa-test-runner/SKILL.md · 在 spec-task 和测试脚本就绪后,逐条执行、分类失败、修最小根因、收集证据、跑 completion 门禁。 你只负责"跑"——不生成报告、不审查代码、不判 Ready。审查留给 qa-code-reviewer,报告与判定留给…
git:20260918.2d0e957 · audit B · 28 stars
qa-test-script-generator skill
mingdui/ming-qa · skills/qa-test-script-generator/SKILL.md · 在用户确认用例之后,把已确认的业务用例展开为可执行的 spec-task 和测试脚本。你产出的是 test-spec-tasks.json——后续执行阶段通过它逐条运行并收集证据。…
git:20260920.56e926f · audit A · 28 stars
qa-testcase-designer skill
mingdui/ming-qa · skills/qa-testcase-designer/SKILL.md · 在风险分析完成后、写测试代码之前,生成中文业务用例并等待用户确认。这是整个 QA 流程中唯一的强制人工门禁——确认后不可回头改用例。 从 context、risk-analysis、已有用例索引出发,生成只含业务行为的…
git:20260918.f90ddaf · audit A · 28 stars

Every file in mingdui/ming-qa

Browse by kind, by grade A, or by owner.