← Back to Skills Marketplace
wangzhiming1999

Bug Investigation

by wangzhiming · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ✓ Security Clean
854
Downloads
0
Stars
5
Active Installs
2
Versions
Install in OpenClaw
/install bug-investigation
Description
Systematically reproduces, locates, and diagnoses frontend bugs using steps, hypotheses, DevTools, and minimal repro. Use when 排查bug, bug定位, 调试, debugging, 复...
README (SKILL.md)

\r \r

Bug 排查(Bug Investigation)\r

\r 用可复现步骤、假设验证和工具链,快速定位并诊断前端问题根因。\r \r

触发场景\r

\r

  • 用户说「排查 bug」「定位问题」「复现不了」「不知道哪里错了」\r
  • 描述现象:白屏、报错、样式错乱、请求失败、交互异常等\r
  • 提供错误信息、截图、环境(浏览器、设备、网络)\r \r

执行流程\r

\r

1. 明确现象与环境\r

\r

  • 现象:用户操作路径 + 预期 vs 实际(一句话描述)\r
  • 环境:浏览器及版本、是否生产/预发/本地、设备/分辨率、网络(慢网/离线)\r
  • 必现/偶现:若偶现,补充可能相关条件(数据量、时机、多 tab 等)\r \r

2. 复现与缩小范围\r

\r

  • 写出最小复现步骤(3–5 步内能复现)\r
  • 判断范围:仅某页面 / 某组件 / 某接口 / 某浏览器 / 某分辨率\r
  • 若无法复现:请用户提供录屏、Console 报错、Network 失败请求,或可复现的测试数据/账号\r \r

3. 假设与验证\r

\r 按优先级做假设并逐条验证(用「是/否」结论):\r \r | 假设类型 | 验证方式 |\r |----------|----------|\r | 前端逻辑/状态 | 断点、console.log、React DevTools 状态 |\r | 接口/数据 | Network 看请求与响应、mock 数据对比 |\r | 样式/布局 | 审查元素、Computed、覆盖样式来源 |\r | 时机/异步 | 执行顺序、Promise/async、事件触发顺序 |\r | 环境/兼容 | 换浏览器、无痕、禁用扩展、不同设备 |\r \r

4. 定位根因\r

\r

  • 指出具体文件 + 位置(行/函数/组件)\r
  • 用一句话说明根因:例如「在 XX 条件下未做判空导致取属性报错」\r
  • 若为接口或后端问题:明确说明「这是后端问题,前端能做的是:1) 加 loading/错误态兜底 2) 给用户友好提示 3) 上报错误日志」,不要继续往前端代码里找\r \r

5. 修复建议\r

\r

  • 给出最小改动修复方案(代码片段或步骤)\r
  • 必要时补充:边界情况、测试建议、类似位置是否需一并检查\r \r

输出模板\r

\r

## Bug 排查报告\r
\r
### 现象\r
- 操作:…\r
- 预期:… / 实际:…\r
\r
### 环境\r
- 浏览器/设备/网络:…\r
\r
### 复现步骤\r
1. …\r
2. …\r
\r
### 假设与验证\r
- [ ] 假设 1:… → 验证结果\r
- [ ] 假设 2:… → 验证结果\r
\r
### 根因\r
- 位置:文件:行 或 组件名\r
- 原因:…\r
\r
### 修复建议\r
- 方案:…\r
- 注意:…\r
```\r
\r
## 常用手段\r
\r
- **Console**:报错栈、`console.trace`、断点\r
- **Network**:状态码、响应体、请求头、是否被缓存\r
- **Elements**:计算样式、覆盖关系、伪元素\r
- **React DevTools**:组件树、props/state、重渲染\r
- **Performance / Lighthouse**:卡顿、长任务、布局抖动\r
- **最小复现**:新建单页或 CodeSandbox 只保留必要代码与数据\r
Usage Guidance
This skill is a harmless, text-only checklist for frontend debugging. Before using it, be careful not to share production secrets, real user credentials, or proprietary source code when supplying recordings, logs, or a minimal repro. If creating public reproductions (CodeSandbox, etc.), sanitize or redact sensitive data. Otherwise the skill is coherent and low-risk.
Capability Analysis
Type: OpenClaw Skill Name: bug-investigation Version: 1.0.1 The skill bundle contains purely procedural instructions and templates for a systematic frontend bug investigation workflow. There is no executable code, shell commands, or network activity, and the instructions in SKILL.md are strictly aligned with the stated purpose of debugging and root cause analysis.
Capability Assessment
Purpose & Capability
Name/description describe frontend bug investigation and the skill only contains procedural debugging instructions; it does not request unrelated credentials, binaries, or system access.
Instruction Scope
Instructions are limited to asking for repro steps, console/network logs, screenshots/recordings, and using DevTools/CodeSandbox; this is appropriate for debugging, but the skill may prompt users to share test accounts, logs, or code—users should avoid sending secrets or proprietary source without review.
Install Mechanism
No install spec or code is present (instruction-only), so nothing is written to disk or fetched during install.
Credentials
No environment variables, credentials, or config paths are requested. The only potential exposure is user-provided repro data (screenshots, logs, test accounts) which is a normal requirement for debugging but should be handled cautiously.
Persistence & Privilege
Skill does not request persistent/always-on presence and uses default invocation settings; it does not modify other skills or system configuration.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install bug-investigation
  3. After installation, invoke the skill by name or use /bug-investigation
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
Publish from oliver-skill repository
v1.0.0
Initial release of bug-investigation skill. - Provides a systematic workflow for reproducing, narrowing down, and diagnosing frontend bugs. - Includes clear steps to clarify the issue, environment, minimal reproduction, hypothesis verification, locating root cause, and proposing fixes. - Offers detailed validation methods for logic, data, style, timing, and compatibility issues. - Supplies an output template for consistent bug investigation reports. - Lists essential debugging tools and tips for effective frontend troubleshooting.
Metadata
Slug bug-investigation
Version 1.0.1
License MIT-0
All-time Installs 5
Active Installs 5
Total Versions 2
Frequently Asked Questions

What is Bug Investigation?

Systematically reproduces, locates, and diagnoses frontend bugs using steps, hypotheses, DevTools, and minimal repro. Use when 排查bug, bug定位, 调试, debugging, 复... It is an AI Agent Skill for Claude Code / OpenClaw, with 854 downloads so far.

How do I install Bug Investigation?

Run "/install bug-investigation" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Bug Investigation free?

Yes, Bug Investigation is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Bug Investigation support?

Bug Investigation is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Bug Investigation?

It is built and maintained by wangzhiming (@wangzhiming1999); the current version is v1.0.1.

💬 Comments