WRAITH

Pure Python Traditional Infrastructure & Web Penetration Testing Framework
The ghost in the wire. No wrappers. No dependencies. Pure engineering.
7
Modules
889
Tests
106
CVEs
500+
Payloads
pip install red-specter-wraith
Docs
PORT SCANNING | SERVICE FINGERPRINTING | SQL INJECTION | XSS | PATH TRAVERSAL | COMMAND INJECTION | SSRF | SSL/TLS ANALYSIS | DEFAULT CREDENTIALS | CMS DETECTION | CVE MAPPING PORT SCANNING | SERVICE FINGERPRINTING | SQL INJECTION | XSS | PATH TRAVERSAL | COMMAND INJECTION | SSRF | SSL/TLS ANALYSIS | DEFAULT CREDENTIALS | CMS DETECTION | CVE MAPPING

Traditional Pentesters Are Wrappers. Real Engineering Is Pure.

Every traditional pentest tool wraps someone else's work. WRAITH doesn't. Every scanner, every fingerprinter, every payload engine — built from scratch in pure Python. No nmap. No sqlmap. No nikto. No wrappers. No excuses.

Wrapper Hell

Every pentest tool wraps nmap, sqlmap, nikto. If the underlying tool changes, breaks, or isn't installed — you're done. Fragile dependency chains everywhere. One update to a wrapped binary and your entire scanner is dead.

No Reasoning

Traditional scanners fire payloads and parse output. They don't understand what they find. They can't chain a SQLi into an LLM injection. They don't think. They don't reason. They just pattern-match and move on.

AI-Blind

Traditional tools don't know AI exists. They can't find an LLM behind a web endpoint. They can't chain traditional vulns into AI exploitation paths. The world moved on. The tools didn't.

Siloed Reports

Port scan results in one tool, web vulns in another, SSL findings in a third. Nothing chains. Nothing correlates. You do the analysis manually. Every. Single. Time.

The WRAITH Armoury

Seven modules. Each one built from scratch in pure Python. No external tool wrappers. No subprocess calls to nmap. No shelling out to sqlmap. Every packet crafted natively. Every payload engine custom-built. 889 tests. 106 CVEs. 500+ payloads.

Module 01

Port Scanner

65,535 Port Coverage

Async TCP connect scanning with banner grabbing. Full 65,535 port coverage. Service detection and response timing analysis. Built on raw sockets. No nmap.

Module 02

Service Fingerprinter

138 Banner Signatures

138 banner signatures with version extraction and confidence scoring. Protocol-aware fingerprinting across SSH, HTTP, FTP, SMTP, MySQL, PostgreSQL, and more.

Module 03

Web Vulnerability Scanner

OWASP Top 10

Full OWASP Top 10 coverage: SQL injection, cross-site scripting, path traversal, command injection, SSRF. Multi-vector payload engine with context-aware injection.

Module 04

SSL/TLS Analyser

Protocol & Cipher Analysis

Protocol version testing, cipher suite enumeration, certificate chain validation, weak configuration detection. Identifies deprecated TLS versions, weak ciphers, and cert issues.

Module 05

Authentication Tester

22 Services

Default credential testing across 22 services. Lockout detection to prevent account lockout. Smart throttling. Covers SSH, FTP, HTTP Basic, databases, admin panels, and more.

Module 06

CMS Detector

14 CMS Types

Detects 14 CMS types including WordPress, Joomla, Drupal, Magento. WordPress plugin enumeration, user enumeration, and known vulnerability mapping per detected version.

Module 07

CVE Checker

106 Real CVEs

106 real CVEs mapped across 22 products. Version-to-CVE mapping with severity scoring. Correlates service fingerprinting results with known vulnerabilities automatically.

One Command. Every Surface.

Target infrastructure, scan every surface:

$ wraith scan --target 192.168.1.0/24 --full --web --cve
[SCAN] Port scanning 254 hosts...
  1,847 open ports found across 23 live hosts
[FINGERPRINT] Identifying services...
  138 signatures matched4 services running outdated versions
[WEB] Scanning web endpoints...
  SQLi confirmed on /api/search — parameter: q
  Reflected XSS on /contact — parameter: name
  Path traversal on /files/download — parameter: path
[SSL] Analysing TLS configurations...
  TLS 1.0 enabled on 3 hosts — deprecated protocol
[CVE] Mapping known vulnerabilities...
  12 CVEs matched3 critical, 5 high, 4 medium
[AUTH] Testing default credentials...
  Default creds found on 2 services — admin:admin

SCAN COMPLETE | 23 hosts | 27 findings | JSON + HTML reports generated

Pure Python

Every module built from scratch. No nmap, no sqlmap, no nikto, no subprocess wrappers. Raw socket scanning, native protocol parsers, custom payload engines.

Correlated Results

Port scans feed service fingerprinting. Fingerprints feed CVE mapping. Web findings chain with authentication results. One unified view. No manual correlation.

NEMESIS Integration

Traditional findings feed directly into NEMESIS. A SQLi found by WRAITH becomes an LLM injection vector. Traditional vulns chain into AI exploitation paths.

AI Shield Integration

Every finding generates an AI Shield blocking rule. Web vulnerabilities that expose AI endpoints become runtime protection policies automatically.

7
Modules
889
Tests
106
CVEs
500+
Payloads
Unleashed Mode

Standard Mode Finds. UNLEASHED Mode Proves.

Standard mode detects vulnerabilities and reports them. UNLEASHED mode proves they're exploitable. Actually extracts database schemas. Actually delivers XSS payloads. Actually reads /etc/passwd. Actually brute-forces authentication. Ed25519 key gate required. Two flags must be passed. This is not accidental.

Capability Standard Unleashed
Port scanning Top 100 ports Full 65535
SQLi Detect injectable parameters Extract schema, dump sample data
XSS Detect reflection Deliver proof-of-concept payloads
Auth Top 50 default creds Full dictionary, extended brute force
Path traversal Detect possibility Read /etc/passwd, .env
Command injection Detect injection point Execute proof commands
CMS Detect version Enumerate users, plugins, known exploits

Ed25519 Gate

UNLEASHED mode requires an Ed25519 private key at ~/.redspecter/override_private.pem and the --override --confirm-destroy flags. Without both, WRAITH operates in detection mode — finding vulnerabilities and documenting what would happen without proving exploitation. The gate is cryptographic. There is no bypass. One key. One operator. Founder's machine only.

The Chain That Changes Everything

WRAITH doesn't just find traditional vulnerabilities. It feeds them into NEMESIS. The Supreme Commander chains traditional findings with AI exploitation in real time. A SQL injection isn't just a SQL injection anymore — it's the first link in a chain that ends with full AI infrastructure compromise.

WRAITH Finds SQLi

Stage 1 — Traditional Discovery

WRAITH discovers a SQL injection on a web endpoint. But behind that endpoint sits an LLM. The traditional vuln is the entry point. WRAITH maps it, confirms it, and passes it forward.

FORGE Injects the LLM

Stage 2 — Model Exploitation

FORGE takes the SQLi entry point and uses it to inject prompts into the LLM behind the endpoint. The traditional vulnerability becomes an AI exploitation vector. The model is compromised.

ARSENAL Exploits the Agent

Stage 3 — Agent Compromise

ARSENAL uses the compromised model to attack the agent layer. Tool manipulation, memory poisoning, goal hijacking. The agent now works for the attacker.

HYDRA Poisons the Supply Chain

Stage 4 — Trust Chain Attack

HYDRA exploits the compromised agent's trust relationships. MCP servers, plugins, marketplaces — the supply chain is poisoned through a trusted agent.

PHANTOM KILL + SCREAMER

Stage 5 — Complete Compromise

PHANTOM KILL owns the foundation layer. SCREAMER blinds the operator. From a single SQL injection to full infrastructure compromise. One chain. Every layer. Nothing assumed safe.

Pure Engineering
Zero External Tools. Zero Wrappers.

Most pentest tools are glorified wrappers around nmap and sqlmap. WRAITH is actual engineering. Every port scanner, every fingerprinter, every payload engine written from scratch in pure Python. Raw sockets. Native protocol parsers. Custom injection engines. No dependencies on tools that can break, change, or disappear.

7
Modules
106
Real CVEs
889
Tests
500+
Payloads

Twelve Tools. Every Layer. No Gaps.

WRAITH is Tool 12 in the Red Specter offensive pipeline. It scans the traditional infrastructure that everything else runs on. Findings feed into NEMESIS for AI exploitation chaining and AI Shield for runtime defence.

Tool 1 — LLM
FORGE
Test the model
Tool 2 — Agent
ARSENAL
Test the AI agent
Tool 3 — Swarm
PHANTOM
Coordinated agent assault
Tool 4 — Web
POLTERGEIST
Test the web layer
Tool 5 — Traffic
GLASS
Watch the wire
Tool 6 — Adversarial
NEMESIS
Think like the attacker
Tool 7 — Human
SPECTER SOCIAL
Target the human
Tool 8 — OS/Kernel
PHANTOM KILL
Own the foundation
Tool 9 — Physical
GOLEM
Attack the physical layer
Tool 10 — Supply Chain
HYDRA
Attack the trust chain
Tool 11 — Operator
SCREAMER
Blind the operator
Tool 12 — Infrastructure
WRAITH
The ghost in the wire
Discovery & Governance
IDRIS
Discovery & governance
Defence Layer
AI SHIELD
Defend everything above
SIEM Integration
redspecter-siem
Enterprise SIEM correlation

Security Distros & Package Managers

Kali Linux
.deb package
Parrot OS
.deb package
BlackArch
PKGBUILD
REMnux
.deb package
Tsurugi
.deb package
PyPI
pip install
Docker
docker-compose

Your Infrastructure Is the Foundation. Test It Pure.

WRAITH scans your traditional infrastructure with zero wrappers and zero dependencies. Pure Python. Every module built from scratch. The ghost in the wire.