Token导航 LogoToken导航

Agent Skill · 研究检索

share-case(分享案例)

share-case 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。

share-case 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 4,642。

总安装 30,862周安装 728GitHub Stars 20下载 10,154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:share-case(分享案例)
来源仓库:https://github.com/haoxuanlithuai/awesome_cognitive_and_neuroscience_skills
仓库路径:skills/share-case
安装命令:
npx skills add https://github.com/haoxuanlithuai/awesome_cognitive_and_neuroscience_skills --skill 'Share Case'
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/haoxuanlithuai/awesome_cognitive_and_neuroscience_skills --skill 'Share Case'

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。

share-case 用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位候选结果。通过 npx skills add 命令安装,需结合原始 README 核验具体用法。安装前建议确认权限范围、维护状态及是否触发联网或文件读写操作。

当前无额外底部简介内容,可参考来源仓库进一步了解功能细节。

来源仓库内路径:skills/share-case

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/haoxuanlithuai/awesome_cognitive_and_neuroscie…;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

share-case 怎么安装?

在终端执行 npx skills add https://github.com/haoxuanlithuai/awesome_cognitive_and_neuroscience_skills…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

share-case 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。

share-case 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。

share-case 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。

share-case 用的人多吗?

公开数据是累计安装 30,862 次,近一周安装 728 次,下载 10,154 次,GitHub 20 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。

share-case 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
share-case

分享案例 · 本页

30,86220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

Share Case

Purpose

This meta-skill captures a researcher's experience using any skill in this repository and submits it as a structured case to GitHub Discussions. It handles all formatting and submission so users never need to leave their terminal or know Git.

When to Use This Skill

Activate when the user:

  • Says "share this case", "share my experience", "分享这个案例"
  • Wants to document how they used a skill in their research
  • Asks how to contribute a usage example

Research Planning Protocol

Before starting the case sharing process, you MUST:

  1. Identify the skill used — Which skill(s) from this repository were used in the current session?
  2. Clarify sharing scope — What aspects of the experience does the user want to share?
  3. Declare what will be extracted — List the specific conversation elements that will be included
  4. Note privacy considerations — Are there dataset names, lab identifiers, or participant details that should be anonymized?
  5. Present the extraction plan to the user and WAIT for confirmation before proceeding.

For detailed methodology guidance, see skills/research-literacy/SKILL.md.

⚠️ Verification Notice

This skill was generated by AI from academic literature. All parameters, thresholds, and citations require independent verification before use in research. If you find errors, please open an issue.


Prerequisites

This skill supports two submission methods:

  1. Direct submission via gh CLI (Recommended)

- Requires gh CLI installed and authenticated - Run gh auth status to check - Install from https://cli.github.com/ and run gh auth login

  1. Manual submission via web browser

- No gh CLI required - Opens GitHub Discussions page in browser - User copies and pastes the generated case

The skill will check for gh availability and let the user choose their preferred method.

GitHub Discussions must be enabled on the repository. The skill submits to the "Show & Tell" category.


Interactive Flow

Step 1 — Quick Survey

Present these questions using multiple-choice format:

Q1: Which skill did you use?

  • Auto-detect from the current session context (list skills that were activated)
  • Let the user confirm or select manually from the full skill list

Q2: What was your research scenario?

  • Formal experiment
  • Coursework or teaching
  • Method exploration
  • Paper reproduction
  • Other (free text)

Q3: How helpful was the skill? (1-5)

  • 1 = Not helpful
  • 2 = Slightly helpful
  • 3 = Moderately helpful
  • 4 = Very helpful
  • 5 = Extremely helpful

Q4: What was most valuable? (select all that apply)

  • Correct methodology guidance
  • Validated parameters and thresholds
  • Pitfall warnings
  • Complete pipeline coverage
  • Literature references
  • Other (free text)

Step 2 — Context Extraction

Extract the following from the current conversation:

  1. Research context — The user's original research question and experimental setup
  2. Key recommendations — The main suggestions the skill provided (parameters, methods, warnings)
  3. Follow-up adjustments — Any modifications made during the conversation
  4. Outcome summary — What the user ultimately decided to do

Format each as a concise paragraph. Do NOT include raw conversation transcripts — synthesize into readable summaries.

Step 3 — User Review and Submission Choice

Display the complete case preview using the format below.

First, check gh CLI availability:

gh auth status 2>&1

Then present submission options using AskUserQuestion:

If gh is available:

  • Submit via gh CLI (Recommended) — Direct submission to GitHub Discussions
  • Open in browser — Manual submission via web interface
  • Delete sections — Remove specific paragraphs before submitting
  • Anonymize — Replace specific names (datasets, labs, participants)
  • Abort — Cancel without submitting

If gh is NOT available:

  • Open in browser — Manual submission via web interface
  • Delete sections — Remove specific paragraphs before submitting
  • Anonymize — Replace specific names (datasets, labs, participants)
  • Abort — Cancel without submitting

You MUST wait for explicit user confirmation before proceeding to Step 4.

Step 4 — Submit to GitHub

Based on the user's choice in Step 3:

Option A: Submit via gh CLI

Create a GitHub Discussion in the "Show & Tell" category.

First, get repository and category IDs:

gh api graphql -f query='
{
  repository(owner: "HaoxuanLiTHUAI", name: "awesome_cognitive_and_neuroscience_skills") {
    id
    discussionCategories(first: 10) {
      nodes {
        id
        name
      }
    }
  }
}'

Then create the discussion:

gh api graphql -f query='
mutation {
  createDiscussion(input: {
    repositoryId: "REPO_ID",
    categoryId: "SHOW_AND_TELL_CATEGORY_ID",
    title: "Community Case: SKILL_NAME",
    body: "CASE_BODY_HERE"
  }) {
    discussion {
      url
    }
  }
}'

On success, display the Discussion URL:

✅ Case shared successfully!
🔗 Discussion URL: [URL]

Thank you for contributing to the community!

On failure, fall back to Option B.

Option B: Open in Browser
  1. Save the case locally to the current directory as case-skill-name-YYYYMMDD.md
  2. Open the GitHub Discussions page:
xdg-open "https://github.com/HaoxuanLiTHUAI/awesome_cognitive_and_neuroscience_skills/discussions/new?category=show-and-tell" 2>/dev/null || \
open "https://github.com/HaoxuanLiTHUAI/awesome_cognitive_and_neuroscience_skills/discussions/new?category=show-and-tell" 2>/dev/null || \
echo "Please open: https://github.com/HaoxuanLiTHUAI/awesome_cognitive_and_neuroscience_skills/discussions/new?category=show-and-tell"
  1. Inform the user:
✅ Case saved to: case-skill-name-YYYYMMDD.md

🌐 Opening GitHub Discussions in your browser...

📋 Next steps:
1. Title: "Community Case: [Skill Name]"
2. Copy the content from case-skill-name-YYYYMMDD.md
3. Paste it into the discussion body
4. Click "Start discussion"

Thank you for contributing to the community!

Case Format Template

The submitted Discussion body uses this format:

## Community Case: [skill-name]

### Quick Info
- **Skill used**: `[skill-name]`
- **Scenario**: [selected option from Q2]
- **Rating**: [stars from Q3, e.g., ⭐⭐⭐⭐]
- **Most valuable**: [selected options from Q4]

### Research Context
> [User's research question and experimental setup — synthesized from conversation]

### What the Skill Suggested
- [Key recommendation 1]
- [Key recommendation 2]
- [Key recommendation 3]

### What I Actually Did & Result
> [User's experience applying the recommendations and the outcome]

### User's Tips
> [Optional: Additional insights the user wants to share with the community]

---
*Submitted via the `share-case` meta-skill.*

Privacy Principles

  • Nothing is submitted without explicit user confirmation after full preview
  • Users can remove any section or paragraph before submission
  • Users can anonymize dataset names, lab names, and participant identifiers
  • Users can choose to submit anonymously (no GitHub username attribution in the case body)
  • The skill never auto-submits — the user must explicitly approve

平台分布

Codex

35.49%
按下载量换算3,604

Claude

30.8%
按下载量换算3,127

Cursor

20.83%
按下载量换算2,115

Gemini CLI

9.11%
按下载量换算925

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