Token导航 LogoToken导航

Agent Skill · 研究检索

qa-generating-bug-reports(qa 生成错误报告)

qa-generating-bug-reports 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

qa-generating-bug-reports 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 71,838。

总安装 192周安装 8GitHub Stars 5下载 64核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:qa-generating-bug-reports(qa 生成错误报告)
来源仓库:https://github.com/wizeline/sdlc-agents
仓库路径:skills/qa-generating-bug-reports
安装命令:
npx skills add https://github.com/wizeline/sdlc-agents --skill qa-generating-bug-reports
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/wizeline/sdlc-agents --skill qa-generating-bug-reports

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

qa-generating-bug-reports 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。它支持基于关键词或任务场景的信息匹配,适用于错误报告生成前的资料准备阶段。

通过安装命令 npx skills add https://github.com/wizeline/sdlc-agents --skill qa-generating-bug-reports 添加技能,具体用法可参考仓库中的 SKILL.md。安装前请确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写操作。

适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/qa-generating-bug-reports

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/wizeline/sdlc-agents --skill qa-generating-bug…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

qa-generating-bug-reports 怎么安装?

在终端执行 npx skills add https://github.com/wizeline/sdlc-agents --skill qa-generating-bug-reports。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

qa-generating-bug-reports 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

qa-generating-bug-reports 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

qa-generating-bug-reports 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

qa-generating-bug-reports 用的人多吗?

公开数据是累计安装 192 次,近一周安装 8 次,下载 64 次,GitHub 5 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

qa-generating-bug-reports 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
qa-generating-bug-reports

qa 生成错误报告 · 本页

1925操作浏览器未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Generating Bug Reports

This skill standardizes bug report creation across all testing agents. Every discovered defect flows through this skill to ensure consistent, actionable, developer-friendly bug reports.

When to Use

  • Any agent discovers a defect during testing
  • You need to format findings for Jira, GitHub Issues, or similar trackers
  • You're compiling a QA report with multiple findings
  • You want to ensure reproduction steps are complete and unambiguous

Bug Report Structure

Every bug report MUST include these sections:

Required Fields

{
  "bug_id": "BUG-001",
  "title": "Short, specific summary (what + where)",
  "severity": "critical|high|medium|low|info",
  "type": "functional|visual|performance|security|ux|accessibility",
  "status": "new",
  "discovered_by": "agent-name",
  "discovered_at": "2025-01-15T10:30:00Z",

  "environment": {
    "url": "https://staging.example.com/checkout",
    "browser": "chromium 120",
    "viewport": "1440x900",
    "os": "Linux",
    "build": "staging-v2.3.1"
  },

  "description": "Full description of what's wrong and why it matters",

  "reproduction_steps": [
    "1. Navigate to /checkout",
    "2. Add item to cart",
    "3. Click 'Proceed to Payment'",
    "4. Leave the email field empty",
    "5. Click 'Submit Order'",
    "6. OBSERVE: Order submits successfully without email validation"
  ],

  "expected_behavior": "Form should show validation error on the email field",
  "actual_behavior": "Order submits with no email, confirmation page shows 'undefined'",

  "evidence": {
    "screenshots": ["screenshot-before.png", "screenshot-after.png"],
    "console_log": "TypeError: Cannot read property 'email' of undefined",
    "network_log": "POST /api/orders → 201 (should have been 400)",
    "har_file": "checkout-session.har"
  },

  "impact": "Users can place orders without email, breaking order confirmation flow",
  "suggested_fix": "Add server-side email validation in POST /api/orders endpoint"
}

Optional Fields

{
  "related_bugs": ["BUG-002", "BUG-005"],
  "affected_users": "All users on checkout flow",
  "frequency": "100% reproducible",
  "workaround": "Manually enter email before submitting",
  "regression": true,
  "regression_since": "v2.3.0",
  "jira_labels": ["checkout", "validation", "regression"],
  "assignee_suggestion": "backend-team"
}

Severity Classification

Use this guide consistently across all agents:

Critical — Application is unusable or data is at risk

  • Authentication bypass or security vulnerability
  • Data loss or corruption
  • Application crashes or is completely non-functional
  • Financial transaction errors (wrong charges, duplicate payments)

High — Major feature is broken with no workaround

  • Core user journey cannot be completed
  • Significant visual regression affecting usability
  • Performance degradation >5x normal
  • API returning incorrect data

Medium — Feature works but with notable issues

  • Non-critical UI elements broken
  • Error messages missing or unhelpful
  • Moderate performance issues (2-5x normal)
  • Cross-browser inconsistencies affecting UX

Low — Minor cosmetic or polish issues

  • Slight visual misalignment (<5px)
  • Minor typos in UI text
  • Inconsistent spacing or styling
  • Deprecated API warnings in console

Info — Observations, not defects

  • Potential improvements
  • Code quality observations
  • Performance optimization opportunities
  • Accessibility suggestions

Writing Good Bug Titles

The title should let a developer understand the issue at a glance:

Pattern: [Area] What happens + Where/When

Good examples:

  • "Checkout: Order submits without email validation on payment form"
  • "Dashboard: Chart data overlaps legend text at tablet viewport"
  • "API: POST /users returns 500 when name exceeds 255 characters"
  • "Login: Password field accepts paste but shows empty on Safari"

Bad examples:

  • "Bug in checkout" (too vague)
  • "Something wrong" (not descriptive)
  • "Form doesn't work" (which form? what doesn't work?)

Writing Reproduction Steps

Reproduction steps should be precise enough that any developer can follow them:

  1. Start from a known state — "Navigate to /checkout" not "Go to the page"
  2. Be specific about inputs — "Enter 'test@example.com'" not "Enter an email"
  3. One action per step — Don't combine "Click login and enter credentials"
  4. Mark the observation point — "OBSERVE:..." or "EXPECTED:... / ACTUAL:..."
  5. Include timing if relevant — "Wait 5 seconds" or "Quickly click 3 times"

Using the Bug Report Generator

python skills/qa-generating-bug-reports/scripts/generate_report.py \
  --template standard \
  --output bugs/BUG-001.json \
  --screenshots screenshot1.png screenshot2.png

Or generate from agent session data:

python skills/qa-generating-bug-reports/scripts/from_session.py \
  --session session-log.json \
  --output bugs/ \
  --agent-name qa-e2e-exploratory-agent

Markdown Report Format

For human-readable reports (e.g., summary documents):

## BUG-001: Checkout order submits without email validation [HIGH]

**Environment:** Chromium 120 / Desktop (1440×900) / staging-v2.3.1
**Discovered:** 2025-01-15 by qa-e2e-exploratory-agent
**Reproducible:** 100%

### Description
When submitting the checkout form with an empty email field, the order
processes successfully. The confirmation page shows "undefined" where
the email should appear.

### Steps to Reproduce
1. Navigate to /checkout
2. Add any item to cart
3. Click 'Proceed to Payment'
4. Leave the email field empty
5. Click 'Submit Order'
6. **OBSERVE:** Order submits, confirmation shows 'undefined' for email

### Expected vs Actual
- **Expected:** Form validation prevents submission, shows error on email field
- **Actual:** Order submits, confirmation page displays "undefined"

### Evidence
- Screenshot: [before-submit.png] [after-submit.png]
- Console: `TypeError: Cannot read property 'email' of undefined`
- Network: `POST /api/orders → 201` (expected 400)

### Impact
All users can place orders without email, breaking confirmation emails
and order tracking.

Deduplication

Before creating a new bug report, check existing reports for duplicates:

  1. Same URL + same error type = likely duplicate
  2. Same console error message = likely duplicate
  3. Same visual region affected = likely duplicate

If a duplicate is found, append the new evidence to the existing report rather than creating a new one. Add a duplicate_of field if merging.

平台分布

Codex

36.44%
按下载量换算23

Claude

30.6%
按下载量换算20

Cursor

15.98%
按下载量换算10

Gemini CLI

8.2%
按下载量换算5

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills