Token导航 LogoToken导航TokenDH.com
研究检索操作浏览器github未标认证来源可访问clear审计未展示

github-issue-creatorGitHub issue creator 开发

Agent Skill

用于围绕 GitHub 仓库、Issue、Pull Request、分支、提交和代码协作流程提供辅助能力。它适合让 Agent 查询项目状态、整理变更、辅助创建或检查协作事项,并把仓库中的信息转成可执行的下一步。使用时需要区分只读查询和写入操作;涉及创建 PR、修改 Issue、推送分支或访问私有仓库时,应确认 token 权限、目标仓库范围和用户授权。

总安装

680

周安装

8

GitHub Stars

公开资料未说明

下载量

65
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

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:github-issue-creator(GitHub issue creator 开发)
来源仓库:https://github.com/edanstarfire/claudecode_webui
仓库路径:skills/github-issue-creator
安装命令:
npx skills add edanstarfire/claudecode_webui --skill "github-issue-creator"
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。

AgentSkills.tonpx skills
npx skills add edanstarfire/claudecode_webui --skill "github-issue-creator"

简介

发现并安装 AI 代理的技能。github-issue-creator 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

  • 适用于多宿主环境技能扩展和生态集成场景。
  • 支持跨 Codex、Claude、Cursor、Gemini CLI 平台部署。
  • 通过 npx 命令直接添加第三方技能仓库。
  • 注意:需验证技能来源可靠性,避免引入安全风险。

SKILL.md

name
github-issue-creator
description
Create well-structured GitHub issues focused on user problems and outcomes, not implementation. Use when user reports bugs or requests features to ensure clear, testable requirements without premature implementation details.
allowed-tools

GitHub Issue Creator

Instructions

When to Invoke This Skill

  • User reports a bug or problem
  • User requests a new feature
  • User suggests quality of life improvement
  • Need to document user need before implementation
  • Converting vague request into clear issue

Core Principle

Issues describe WHAT and WHY, not HOW.

Issues are contracts between user needs and development outcomes. They should enable:

  • Clear understanding of the problem/need
  • Objective evaluation of completeness
  • Testing whether solution succeeds
  • Flexibility in implementation approach

Separate concerns:

  • Issue = User problem, expected outcome, acceptance criteria
  • Implementation Plan = Technical approach (comes later, in comments or design docs)
  • PR Description = What was actually implemented
  • Code Comments = Why specific code decisions were made

Issue Structure

Template

## Problem / Need

<Clear description of what's wrong or what's needed from user perspective>

### Current Behavior (for bugs)
<What happens now that shouldn't>

### Expected Behavior
<What should happen instead, from user's point of view>

## Impact

### User Impact
<Who is affected and how?>
<What can't users do? What's frustrating?>

### Business/Project Impact
<Why does this matter?>
<What's the cost of not fixing this?>

## User Perspective

### User Story (optional but recommended)
As a [type of user]
I want to [action/capability]
So that [benefit/outcome]

### User Journey
1. User does X
2. [Current: System does Y / Expected: System should do Z]
3. User experiences [problem/benefit]

## Quality Standards

### Backward Compatibility
- [ ] Must not break existing functionality
- [ ] Can break existing functionality (explain why)
- [ ] New feature (no compatibility concern)

### Data Persistence
- [ ] Changes must persist across restarts
- [ ] Changes must persist across sessions
- [ ] Transient state (no persistence needed)

### Performance
- [ ] No performance requirements
- [ ] Must complete within [time]
- [ ] Must handle [scale] concurrent users/operations

### Security
- [ ] No security implications
- [ ] Requires authentication/authorization
- [ ] Handles sensitive data

## Acceptance Criteria

### Definition of Done
- [ ] <Testable outcome 1>
- [ ] <Testable outcome 2>
- [ ] <Testable outcome 3>

### Test Scenarios
1. **Scenario:** <Description>
   - **Given:** <Initial state>
   - **When:** <User action>
   - **Then:** <Expected result>

2. **Scenario:** <Another scenario>
   - **Given:** <Initial state>
   - **When:** <User action>
   - **Then:** <Expected result>

### Edge Cases to Consider
- <Edge case 1>
- <Edge case 2>

### Out of Scope (for this issue)
- <Related feature that's NOT included>
- <Future enhancement that's separate>

## Additional Context

### Related Issues
- Fixes #<issue>
- Related to #<issue>
- Blocks #<issue>

### References
- <Links to user feedback>
- <Links to documentation>
- <Screenshots/recordings>

### Notes
<Any additional context that helps understand the need>

---

## ⚠️ Implementation Details

**Implementation planning happens AFTER issue is approved.**

Once this issue is accepted:
1. Create implementation plan in comments or design doc
2. Break down into technical tasks
3. Identify files/functions to modify
4. Plan testing approach
5. Execute implementation

**Do NOT include in initial issue:**
- Specific files to modify
- Specific function names
- Exact variable names
- Detailed code structure
- UI mockups (unless critical to understanding the need)

Conversation Flow to Create Issue

1. Extract User Need

Ask clarifying questions:

  • What are you trying to accomplish?
  • What happens currently?
  • What should happen instead?
  • Who is affected by this?
  • Why is this important?

Avoid asking:

  • How should we implement this? (too early)
  • Which files should we change? (not user concern)
  • What should we name variables? (implementation detail)

2. Understand Impact

Explore:

  • How often does this happen?
  • How severe is the impact?
  • Who experiences this?
  • What's the workaround (if any)?
  • What happens if we don't fix this?

3. Define Expected Behavior

Focus on observable outcomes:

  • ✅ "User should see error message explaining what went wrong"
  • ❌ "Display error in a red div with class 'error-message'"
  • ✅ "Data should persist across browser sessions"
  • ❌ "Store data in localStorage using JSON.stringify()"
  • ✅ "Response time should be under 2 seconds"
  • ❌ "Implement caching using Redis"

4. Establish Acceptance Criteria

Use Given-When-Then format:

Given: <starting state>
When: <user action>
Then: <observable result>

Make criteria testable:

  • ✅ "Submit button is disabled when form is invalid"
  • ❌ "Form validation should work properly"
  • ✅ "Clicking 'Delete' shows confirmation dialog before deletion"
  • ❌ "Add confirmation for delete"

5. Identify Quality Standards

Ask about:

  • Can this break existing features?
  • Does data need to persist?
  • Are there performance requirements?
  • Are there security concerns?

6. Clarify Scope

What's included:

  • Core functionality needed

What's excluded (out of scope):

  • Nice-to-haves for future
  • Related but separate features
  • Edge cases that can wait

Issue Quality Checklist

Before creating the issue, verify:

Problem/Need is Clear:

  • [ ] Anyone can understand what's wrong or needed
  • [ ] Current vs expected behavior is obvious
  • [ ] Context is sufficient

Impact is Explained:

  • [ ] Who is affected
  • [ ] Why it matters
  • [ ] Severity/priority is justified

User Perspective is Maintained:

  • [ ] Written from user's viewpoint
  • [ ] Focuses on outcomes, not internals
  • [ ] Describes behavior, not code

Acceptance Criteria are Testable:

  • [ ] Each criterion is objective
  • [ ] Can verify each with test
  • [ ] No ambiguous terms ("better", "improved", "nice")

Quality Standards are Defined:

  • [ ] Backward compatibility addressed
  • [ ] Data persistence specified
  • [ ] Performance needs stated (if any)

Implementation Details are Absent:

  • [ ] No specific file names
  • [ ] No function/class names
  • [ ] No UI mockups (unless essential)
  • [ ] No technical architecture

Scope is Bounded:

  • [ ] Focus is narrow enough
  • [ ] Out of scope is documented
  • [ ] Related issues are linked

Creating the Issue

Command:

gh issue create --title "<type>: <brief description>" --body "$(cat <<'EOF'
<issue content using template above>
EOF
)"

Title Format:

  • feat: <user-facing feature>
  • fix: <problem being fixed>
  • chore: <maintenance task>
  • docs: <documentation improvement>
  • perf: <performance improvement>

Title Guidelines:

  • Brief (50 chars or less)
  • User-focused ("Add dark mode toggle")
  • Not implementation-focused ("Refactor CSS variables")
  • Descriptive without being technical

Common Anti-Patterns to Avoid

❌ Anti-Pattern 1: Implementation in Disguise

## Problem
We need to add a new endpoint POST /api/users/avatar

## Solution
1. Add route in web_server.py
2. Create upload handler in storage.py
3. Update User model with avatar_url field

Why it's wrong: This is an implementation plan, not a user need.

Better version:

## Problem
Users cannot upload profile avatars

### Current Behavior
User profiles display only default avatar

### Expected Behavior
Users can upload custom avatar image that displays on their profile

## Acceptance Criteria
- User can select image file from device
- Uploaded avatar displays immediately
- Avatar persists across sessions
- Avatar size is limited to 2MB

❌ Anti-Pattern 2: Vague Requirements

## Problem
Make the UI better

## Expected
Better user experience

Why it's wrong: "Better" is subjective, not testable.

Better version:

## Problem
Users struggle to find the delete button for sessions

### Current Behavior
Delete button only appears on hover and users miss it

### Expected Behavior
Delete button should be visible at all times for easy access

### User Impact
Users accidentally accumulate old sessions, cluttering their workspace

## Acceptance Criteria
- Delete button is visible without hover
- Button placement is consistent across all session items
- Button has clear "Delete" label or icon

❌ Anti-Pattern 3: Mixed Concerns

## Feature Request
Add dark mode and refactor CSS to use variables and update
documentation and add unit tests for theme switching

Why it's wrong: Multiple separate concerns bundled together.

Better version: Create separate issues:

  1. feat: Add dark mode toggle for user interface
  2. chore: Update UI documentation for theming
  3. (CSS refactoring happens during implementation, not separate issue)
  4. (Tests are part of acceptance criteria, not separate issue)

❌ Anti-Pattern 4: Solution Instead of Problem

## Feature Request
Add a Redis cache for session data

Why it's wrong: Proposes solution without explaining problem.

Better version:

## Problem
Session list takes 5+ seconds to load when user has 100+ sessions

### Current Behavior
Every page load fetches all session data from disk
Page is unresponsive during loading

### User Impact
Users with many sessions experience slow, frustrating interface
Unable to work efficiently

## Acceptance Criteria
- Session list loads in under 1 second for 100 sessions
- UI shows loading indicator during fetch
- List remains responsive during load

## Performance Requirement
Support up to 500 sessions per user with sub-2-second load times

---
Note: Redis, caching, pagination, or other solutions can be explored
during implementation planning

Examples

Example 1: Bug Report

## Problem

WebSocket connection fails to reconnect after network interruption

### Current Behavior
When user's network connection drops (WiFi disconnect, laptop sleep):
- WebSocket connection is lost
- Messages stop flowing
- UI shows "connected" status incorrectly
- User must refresh page to reconnect

### Expected Behavior
When network connection is restored:
- WebSocket automatically reconnects
- UI shows "reconnecting" status
- Messages resume flowing
- No page refresh required

## Impact

### User Impact
Users lose work when network drops and don't realize connection is broken.
Must manually refresh and potentially lose unsent messages.

### Frequency
Common for laptop users who move between locations or close laptop lid.

## Acceptance Criteria

- [ ] Connection automatically reconnects when network restored
- [ ] UI accurately reflects connection state (connected/reconnecting/disconnected)
- [ ] Unsent messages are queued and sent after reconnection
- [ ] No data loss during reconnection

### Test Scenarios

1. **Scenario:** Network interruption during active session
   - **Given:** User has active session with WebSocket connected
   - **When:** Network disconnects then reconnects after 10 seconds
   - **Then:** WebSocket reconnects automatically, queued messages sent

2. **Scenario:** Laptop sleep/wake cycle
   - **Given:** User has active session
   - **When:** User closes laptop lid then reopens after 1 hour
   - **Then:** WebSocket reconnects, session resumes without refresh

Example 2: Feature Request

## Problem / Need

Users cannot organize sessions into folders or categories

### Current Situation
All sessions appear in flat list within project.
Users with 20+ sessions struggle to find specific session.

## Impact

### User Impact
Users waste time scrolling through long session lists.
No way to separate active vs archived sessions.

### User Story
As a developer managing multiple features,
I want to organize sessions into folders,
So that I can quickly find sessions related to specific work areas.

## Quality Standards

### Backward Compatibility
- [ ] Must support existing flat session structure

### Data Persistence
- [ ] Folder structure must persist across restarts

## Acceptance Criteria

- [ ] Users can create named folders within projects
- [ ] Users can move sessions between folders
- [ ] Users can rename folders
- [ ] Users can delete empty folders
- [ ] Folder structure persists across browser sessions

### Out of Scope
- Folder-level settings (separate feature)
- Sharing folders between projects

Summary

This skill creates issues that clearly describe user problems and expected outcomes without constraining implementation approach, enabling flexible technical solutions while maintaining clear success criteria.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

windsurf

29.41%
按下载量换算19

OpenCode

20.52%
按下载量换算13

Codex

19.84%
按下载量换算13

Claude Code

11.49%
按下载量换算7

Antigravity

7.23%
按下载量换算5

Gemini CLI

3.52%
按下载量换算2

安全审计

暂无安全审计结果可展示。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills