Token导航 LogoToken导航

Agent Skill · 研究检索

pre-work-research(前期工作研究)

pre-work-research 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。

pre-work-research 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,权限为调用外部服务;许可证为 MIT,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 56,510。

总安装 449周安装 18GitHub Stars 6下载 145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
MIT

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:pre-work-research(前期工作研究)
来源仓库:https://github.com/troykelly/claude-skills
仓库路径:skills/pre-work-research
安装命令:
npx skills add https://github.com/troykelly/claude-skills --skill pre-work-research
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/troykelly/claude-skills --skill pre-work-research

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。

pre-work-research 用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词、任务场景或来源线索快速定位候选结果。通过 npx skills add 命令从指定 GitHub 仓库安装并使用。

安装前需确认权限范围和维护状态,注意可能触发联网、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/pre-work-research

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/troykelly/claude-skills --skill pre-work-resea…;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

    它需要调用外部服务,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

pre-work-research 怎么安装?

在终端执行 npx skills add https://github.com/troykelly/claude-skills --skill pre-work-research。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

pre-work-research 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。

pre-work-research 需要什么权限?

权限级别是调用外部服务。该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

pre-work-research 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

pre-work-research 用的人多吗?

公开数据是累计安装 449 次,近一周安装 18 次,下载 145 次,GitHub 6 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。

pre-work-research 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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
pre-work-research

前期工作研究 · 本页

4496调用外部服务MIT有提醒

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

Pre-Work Research

Overview

Research before coding. Understand the landscape before changing it.

Core principle: Measure twice, cut once.

Announce at start: "I'm researching the codebase and documentation before implementing."

When to Research

Research is appropriate when:

SituationResearch Needed
Unfamiliar area of codebaseYes
New library or APIYes
Complex integrationYes
Performance-sensitive codeYes
Security-sensitive codeYes
Simple, isolated changeMinimal

The Research Protocol

Step 1: Understand Requirements

Before researching implementation:

  • Read the full issue description
  • Review all acceptance criteria
  • Note any constraints mentioned
  • Identify unknowns

Step 2: Research Repository Documentation

# Project overview
cat README.md

# Architecture/design docs
ls -la docs/
find . -name "*.md" -path "./docs/*"

# Contributing guidelines
cat CONTRIBUTING.md

# API documentation
cat docs/api*.md
cat docs/architecture*.md

Look for:

  • Architecture decisions
  • Design patterns used
  • Coding conventions
  • Testing requirements
  • Deployment considerations

Step 3: Research Existing Codebase

Find Similar Implementations
# Find similar features
grep -r "similar keyword" src/

# Find related tests
grep -r "similar keyword" **/*.test.*

# Find imports of relevant modules
grep -r "import.*ModuleName" src/
Understand Patterns
# How are similar things done?
# Look at 2-3 examples of similar functionality

# Example: If adding an API endpoint
grep -r "router\." src/routes/
cat src/routes/[existing-endpoint].ts

# Example: If adding a service
ls -la src/services/
cat src/services/[existing-service].ts
Check for Utilities
# What utilities exist?
ls -la src/utils/
cat src/utils/[relevant-util].ts

# Are there shared helpers?
grep -r "export function" src/utils/

Step 4: Research External Resources

When using external APIs or libraries:

Official Documentation
# Read the docs for dependencies
pnpm info [package-name]
# Then visit documentation URL
Search: "[library-name] documentation"
Search: "[library-name] getting started"
Search: "[library-name] [specific feature]"
API References

For external APIs:

  • Authentication requirements
  • Rate limits
  • Error handling
  • Response formats
  • Versioning
Community Resources
  • GitHub issues for common problems
  • Stack Overflow for patterns
  • Blog posts for best practices

Step 5: Document Findings

Create a research summary:

## Pre-Work Research: Issue #[NUMBER]

### Requirements Understanding
- [Key requirement 1]
- [Key requirement 2]

### Codebase Patterns
- Pattern for [X]: See `src/example/pattern.ts`
- Utilities available: `src/utils/helper.ts`
- Test pattern: See `src/example/example.test.ts`

### External Dependencies
- [Library]: [Key findings]
- [API]: [Authentication method, rate limits]

### Approach
Based on research, the approach is:
1. [Step 1]
2. [Step 2]
3. [Step 3]

### Risks/Considerations
- [Risk 1]
- [Consideration 1]

Step 6: Update Issue (If Significant)

If research reveals important context:

gh issue comment [ISSUE_NUMBER] --body "## Pre-Implementation Research

### Approach
[Summary of planned approach]

### Considerations
- [Important finding 1]
- [Important finding 2]

### Questions (if any)
- [Question needing clarification]
"

Research Depth by Task Size

Task SizeResearch Depth
Trivial (typo, config)None needed
Small (single file)Quick pattern check
Medium (feature)Full protocol
Large (system)Extended research

Quick Pattern Check (5 min)

# Just verify pattern
grep -r "pattern" src/ | head -5
cat src/similar/example.ts | head -50

Full Protocol (15-30 min)

Complete Steps 1-6 above.

Extended Research (1+ hour)

  • Read all relevant documentation
  • Trace through existing implementations
  • Create proof-of-concept if needed
  • Document architectural considerations

What to Look For

In Documentation

Look ForWhy
Architecture diagramsUnderstand system structure
Coding standardsMatch existing style
Decision recordsUnderstand why things are done a way
API contractsMaintain compatibility

In Codebase

Look ForWhy
Similar featuresFollow established patterns
Test patternsWrite consistent tests
Error handlingHandle errors consistently
Logging patternsLog appropriately

In External Resources

Look ForWhy
Official examplesUse recommended patterns
Common pitfallsAvoid known issues
Performance tipsOptimize appropriately
Security guidanceImplement securely

Research Outputs

After research, you should know:

  • How similar features are implemented
  • What patterns to follow
  • What utilities are available
  • What the testing approach should be
  • Any risks or special considerations

Checklist

Before starting implementation:

  • Issue requirements understood
  • Repository docs checked
  • Similar code patterns found
  • Relevant utilities identified
  • External resources researched (if applicable)
  • Approach documented
  • Issue updated (if significant findings)

Integration

This skill is called by:

  • issue-driven-development - Step 5

This skill informs:

  • tdd-full-coverage - How to write tests
  • strict-typing - Type patterns to use
  • inline-documentation - Documentation patterns

平台分布

Claude Code

27.59%
按下载量换算40

Antigravity

23.1%
按下载量换算33

Gemini CLI

18.72%
按下载量换算27

OpenCode

12.13%
按下载量换算18

Cursor

7.11%
按下载量换算10

kiro-cli

3.5%
按下载量换算5

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

调用外部服务

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。

继续浏览同类 Skills