Token导航 LogoToken导航

Agent Skill · 研究检索

conducting-social-engineering-penetration-test(进行社会工程渗透测试)

用于辅助测试设计、自动化测试、用例整理和回归验证。它适合让 Agent 编写单元测试、端到端测试、测试计划或根据失败日志定位问题。使用时需要确认项目测试框架、运行命令和夹具数据,避免为了通过测试而改坏真实逻辑;涉及浏览器或外部服务时,应区分本地模拟、测试环境和生产环境。

conducting-social-engineering-penetration-test 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,权限为需要联网;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 52,984。

总安装 541周安装 23GitHub Stars 5,869下载 190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:conducting-social-engineering-penetration-test(进行社会工程渗透测试)
来源仓库:https://github.com/mukul975/anthropic-cybersecurity-skills
仓库路径:skills/conducting-social-engineering-penetration-test
安装命令:
npx skills add https://github.com/mukul975/anthropic-cybersecurity-skills --skill conducting-social-engineering-penetration-test
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/mukul975/anthropic-cybersecurity-skills --skill conducting-social-engineering-penetration-test

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。

用于模拟社会工程攻击,评估组织对欺骗手段的抵御能力。覆盖钓鱼、语音诈骗(vishing)、短信诈骗(smishing)和 pretexting 等多种方式。结合 GoPhish、SET 等工具执行可控测试,衡量员工安全意识水平。仅限授权范围内开展,禁止用于非法目的或未经同意的人员接触。

conducting-social-engineering-penetration-test 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/conducting-social-engineering-penetration-test

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/mukul975/anthropic-cybersecurity-skills --skil…;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

    它需要需要联网,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

conducting-social-engineering-penetration-test 怎么安装?

在终端执行 npx skills add https://github.com/mukul975/anthropic-cybersecurity-skills --skill conducti…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

conducting-social-engineering-penetration-test 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。

conducting-social-engineering-penetration-test 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

conducting-social-engineering-penetration-test 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。

conducting-social-engineering-penetration-test 用的人多吗?

公开数据是累计安装 541 次,近一周安装 23 次,下载 190 次,GitHub 5,869 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。

conducting-social-engineering-penetration-test 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。
  • 来源仓库有 5,869 星,维护活跃度相对可查。

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
conducting-social-engineering-penetration-test

进行社会工程渗透测试 · 本页

5415,869需要联网未明确未通过

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

Conducting Social Engineering Penetration Test

Overview

Social engineering penetration testing assesses an organization's human attack surface through controlled simulation of real-world deception techniques. According to Verizon DBIR 2024, the human element is involved in approximately 68% of all breaches, with phishing remaining the dominant initial access vector. This skill covers phishing, vishing (voice phishing), smishing (SMS phishing), and physical pretexting campaigns using tools like GoPhish, the Social Engineer Toolkit (SET), and Evilginx.

When to Use

  • When conducting security assessments that involve conducting social engineering penetration test
  • When following incident response procedures for related security events
  • When performing scheduled security testing or auditing activities
  • When validating security controls through hands-on testing

Prerequisites

  • Written authorization from senior management (CISO/CTO)
  • Legal review confirming compliance with local laws (CFAA, GDPR, etc.)
  • Defined scope: target employee groups, attack types, exclusions
  • GoPhish server, domain for phishing infrastructure, VPS
  • OSINT tools: Maltego, theHarvester, LinkedIn scraping tools
  • Coordination with HR and Legal for employee notification post-test

Phase 1 — OSINT and Target Profiling

Employee Reconnaissance

# Email harvesting
theHarvester -d targetcorp.com -b all -l 500 -f harvester_results

# LinkedIn OSINT (manual + tools)
# Gather: names, roles, departments, email format
# Identify: new hires, IT admins, finance team, executives

# Email format discovery
# Check: first.last@, flast@, firstl@
# Verify with: hunter.io, phonebook.cz, email-checker.net

# Social media OSINT
# Twitter/X: employees posting about work tools/frustrations
# Facebook: corporate event photos, office layout
# GitHub: employee repos with corporate email addresses

# Domain intelligence
dig targetcorp.com MX +short
dig targetcorp.com TXT +short
# Check for SPF, DKIM, DMARC records
# Weak DMARC = easier email spoofing

# Check breach databases (authorized)
# HaveIBeenPwned API for corporate emails
# Identify employees with compromised credentials

Target Selection Matrix

GroupCountPretextAttack Vector
Finance15Invoice approvalPhishing (credential harvest)
IT Help Desk8Password resetVishing
Executives5Board meeting updateSpear phishing
New Hires (< 90 days)12HR onboarding formPhishing (payload)
All Employees200IT security updateBroad phishing

Phase 2 — Phishing Campaign

Infrastructure Setup

# Register lookalike domain
# targetcorp.com -> targetc0rp.com, targetcorp-secure.com, targetcorp.net

# Set up GoPhish server
sudo apt install gophish
# Edit config.json for admin and phishing server ports
gophish

# Configure sending profile in GoPhish
# SMTP server: mail.attackdomain.com
# From: it-security@targetcorp-secure.com
# Headers: proper DKIM/SPF for attack domain

# Create landing page (credential harvesting)
# Clone legitimate login page (Office 365, Okta, etc.)
# GoPhish: Import Site -> https://login.microsoftonline.com
# Enable: Capture Credentials, Capture Passwords
# Redirect to real site after capture

# Set up Evilginx for MFA bypass (authorized testing only)
evilginx2
: config domain attackdomain.com
: config ipv4 <server_ip>
: phishlets hostname o365 login.targetcorp-secure.com
: phishlets enable o365
: lures create o365
: lures get-url 0

GoPhish Campaign Configuration

{
  "campaign": {
    "name": "IT Security Update - Q1 2025",
    "template": {
      "name": "Mandatory Security Training",
      "subject": "Action Required: Complete Security Awareness Training by Friday",
      "html": "<html>...[branded email with urgency]...</html>",
      "from": "IT Security Team <security@targetcorp-secure.com>"
    },
    "landing_page": "Office 365 Login Clone",
    "sending_profile": "Phishing SMTP",
    "groups": ["All Employees - Batch 1"],
    "launch_date": "2025-03-10T09:00:00Z",
    "send_by_date": "2025-03-10T12:00:00Z"
  }
}

Phishing Email Templates by Pretext

Template 1 — IT Security Update:

Subject: [Action Required] Mandatory Password Reset - Security Incident
From: IT Security <security@targetcorp-secure.com>

Dear {FirstName},

Our security team has detected unauthorized access attempts on our systems.
As a precautionary measure, all employees must reset their passwords immediately.

Please click below to reset your password within the next 24 hours:

[Reset Password Now] -> {phishing_url}

Failure to comply may result in temporary account suspension.

Thank you,
IT Security Team

Template 2 — Finance Invoice:

Subject: Invoice #INV-2025-4821 - Approval Required
From: Accounts Payable <ap@targetcorp-secure.com>

Hi {FirstName},

Please review and approve the attached invoice from our vendor.
Amount: $47,250.00 | Due: March 15, 2025

[View Invoice] -> {phishing_url}

Best regards,
Accounts Payable

Phase 3 — Vishing Campaign

Call Script Template

Pretext: IT Help Desk calling about suspicious login

Caller: "Hi, this is [Name] from the IT Help Desk. Am I speaking with [Target Name]?"

[Wait for confirmation]

Caller: "We've detected some unusual login activity on your account from an
unrecognized location. For your protection, I need to verify your identity
before we can investigate further."

Caller: "Can you confirm your employee ID and the email address associated
with your account?"

[Record responses]

Caller: "Thank you. I'm going to send you a verification link to confirm
it's really you. Can you click on it and enter your credentials so we can
secure your account?"

[Send phishing link via email/SMS during call]

Caller: "Great, I can see you've been verified. Your account is now secured.
If you notice any further issues, please call the help desk at [real number]."

Vishing Metrics to Track

MetricDescription
Call answeredTarget picked up the phone
EngagedTarget continued conversation past initial question
Information disclosedTarget provided credentials, employee ID, or PII
Link clickedTarget clicked the verification link
Credentials enteredTarget entered credentials on phishing page
ReportedTarget reported the call to security

Phase 4 — Physical Social Engineering

Physical Pretexting Scenarios

Scenario 1: Delivery Person
- Arrive with package labeled for executive
- Request access to deliver personally
- Attempt to tailgate through secure doors
- Drop USB drives in common areas

Scenario 2: IT Vendor
- Arrive with vendor badge (printed)
- Claim scheduled maintenance on network closet
- Attempt to access server rooms
- Install rogue wireless AP if access gained

Scenario 3: New Employee
- Arrive claiming first day orientation
- Request temporary badge
- Attempt to access restricted areas
- Photograph sensitive screens/documents

Evidence Collection:
- Body camera (if legally permitted and authorized)
- Photographs of accessed areas
- WiFi probe from rogue AP
- Notes on which doors/checkpoints bypassed

Phase 5 — Metrics and Analysis

Campaign Results Dashboard

Phishing Campaign Results:
├── Emails Sent: 200
├── Emails Delivered: 195 (97.5%)
├── Emails Opened: 142 (72.8%)
├── Links Clicked: 68 (34.9%)
├── Credentials Submitted: 31 (15.9%)
├── MFA Bypassed: 8 (4.1%)  [Evilginx]
├── Reported to SOC: 12 (6.2%)
└── No Action: 53 (27.2%)

Vishing Campaign Results:
├── Calls Made: 23
├── Calls Answered: 18 (78.3%)
├── Engaged in Conversation: 15 (65.2%)
├── Information Disclosed: 9 (39.1%)
├── Credentials Provided: 4 (17.4%)
└── Reported to Security: 2 (8.7%)

Physical Assessment:
├── Tailgating Successful: 3/5 attempts
├── USB Drives Plugged In: 2/10 dropped
├── Restricted Areas Accessed: 2/4 attempted
└── Challenged by Employee: 1 time

Risk Scoring

Attack VectorSuccess RateRisk LevelPriority
Phishing (credential harvest)15.9%HighP1
Vishing (info disclosure)39.1%CriticalP1
Physical tailgating60%HighP2
USB drop20%MediumP3
Spear phishing (exec)40%CriticalP1

Phase 6 — Reporting and Recommendations

Remediation Priorities

PriorityRecommendationTimeline
P1Deploy phishing-resistant MFA (FIDO2/WebAuthn)30 days
P1Implement targeted security awareness training14 days
P1Deploy email gateway with URL rewriting30 days
P2Strengthen physical access controls (mantraps, visitor badges)60 days
P2Implement security champion program per department30 days
P3Deploy USB device control policy30 days
P3Establish phishing reporting button in email client14 days

Tools Reference

ToolPurpose
GoPhishPhishing campaign management platform
Evilginx2MFA bypass via reverse proxy phishing
Social Engineer Toolkit (SET)Social engineering attack framework
MaltegoOSINT and relationship mapping
theHarvesterEmail and domain OSINT
King PhisherPhishing campaign tool
ModlishkaReverse proxy for credential interception

References

平台分布

Codex

34.46%
按下载量换算65

Claude

28.62%
按下载量换算54

Cursor

19.71%
按下载量换算37

Gemini CLI

10.73%
按下载量换算20

安全审计

Gen Agent Trust Hub

通过

Socket

未通过

Snyk

未通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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