First commit

.dockerignore

@@ -0,0 +1,14 @@
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.Python
+*.so
+*.egg
+*.egg-info
+dist
+build
+.git
+.gitignore
+README.md
+*.md

.gitignore

@@ -0,0 +1,71 @@
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# Virtual Environment
+venv/
+env/
+ENV/
+env.bak/
+venv.bak/
+
+# IDEs
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+.DS_Store
+
+# Helm
+helm/charts/
+helm/*.tgz
+helm/values-production.yaml
+helm/values-*.yaml
+!helm/values.yaml
+
+# Kubernetes secrets (if generated locally)
+*.secret.yaml
+secrets/
+
+# Docker
+*.log
+
+# Environment variables
+.env
+.env.local
+.env.*.local
+
+# Logs
+*.log
+logs/
+
+# Temporary files
+*.tmp
+*.temp
+.cache/
+
+# Personal canary tokens or sensitive configs
+*canary*token*.yaml
+personal-values.yaml

Dockerfile

@@ -0,0 +1,19 @@
+FROM python:3.11-slim
+
+LABEL org.opencontainers.image.source=https://github.com/BlessedRebuS/Krawl
+
+WORKDIR /app
+
+COPY src/ /app/src/
+COPY wordlists.json /app/
+
+RUN useradd -m -u 1000 krawl && \
+    chown -R krawl:krawl /app
+
+USER krawl
+
+EXPOSE 5000
+
+ENV PYTHONUNBUFFERED=1
+
+CMD ["python3", "src/server.py"]

README.md

@@ -1,2 +1,297 @@
-# Krawl
+<h1 align="center">🕷️ Krawl</h1>
-Krawl is a Web Honeypot &amp; Deception server that aims to foul enumerations, web crawling, fuzzing and bruteforcing
+
+<p align="center">
+  A modern, customizable zero-dependencies honeypot server designed to detect and track malicious activity through deceptive web pages, fake credentials, and canary tokens.
+</p>
+
+<div align="center">
+  <a href="https://github.com/blessedrebus/krawl/blob/main/LICENSE">
+    <img src="https://img.shields.io/github/license/blessedrebus/krawl" alt="License">
+  </a>
+  <a href="https://github.com/blessedrebus/krawl/releases">
+    <img src="https://img.shields.io/github/v/release/blessedrebus/krawl" alt="Release">
+  </a>
+</div>
+
+<div align="center">
+  <a href="https://ghcr.io/blessedrebus/krawl">
+    <img src="https://img.shields.io/badge/ghcr.io-krawl-blue" alt="GitHub Container Registry">
+  </a>
+  <a href="https://kubernetes.io/">
+    <img src="https://img.shields.io/badge/kubernetes-ready-326CE5?logo=kubernetes&logoColor=white" alt="Kubernetes">
+  </a>
+  <a href="https://github.com/BlessedRebuS/Krawl/pkgs/container/krawl-chart">
+    <img src="https://img.shields.io/badge/helm-chart-0F1689?logo=helm&logoColor=white" alt="Helm Chart">
+  </a>
+</div>
+
+<br>
+
+<p align="center">
+  <a href="#-overview">Overview</a> •
+  <a href="#-quick-start">Quick Start</a> •
+  <a href="#%EF%B8%8F-configuration">Configuration</a> •
+  <a href="#-dashboard">Dashboard</a> •
+  <a href="#-deception-techniques">Deception Techniques</a> •
+  <a href="#-contributing">Contributing</a>
+</p>
+
+![asd](img/deception-page.png)
+
+## What is Krawl?
+
+Krawl is a simple cloud native deception server that creates fake web applications with low hanging fruit and juicy fake random information. 
+
+It features:
+
+- **Spider Trap Pages**: Infinite random links to waste crawler resources based on the [spidertrap project](https://github.com/adhdproject/spidertrap)
+- **Fake Login Pages**: WordPress, phpMyAdmin, admin panels
+- **Honeypot Paths**: Advertised in robots.txt to catch scanners
+- **Fake Credentials**: Realistic-looking usernames, passwords, API keys
+- **Canary Token Integration**: External alert triggering
+- **Real-time Dashboard**: Monitor suspicious activity
+- **Customizable Wordlists**: Easy JSON-based configuration
+- **Random Error Injection**: Mimic real server behavior
+
+## 🚀 Quick Start
+## Helm Chart
+
+Install with default values
+
+```bash
+helm install krawl ./helm \
+  --namespace krawl-system \
+  --create-namespace
+```
+
+Install with custom values
+
+```bash
+helm install krawl ./helm \
+  --namespace krawl-system \
+  --create-namespace \
+  --values values.yaml
+```
+
+Install with custom canary token
+
+```bash
+helm install krawl ./helm \
+  --namespace krawl-system \
+  --create-namespace \
+  --set config.canaryTokenUrl="http://your-canary-token-url"
+```
+
+Uninstall with
+```bash
+helm uninstall krawl --namespace krawl-system
+```
+
+## Kubernetes / Kustomize
+Apply all manifests
+
+```bash
+kubectl apply -k manifests/
+```
+Retrieve dashboard path
+```bash
+kubectl get secret krawl-server -n krawl-system -o jsonpath='{.data.dashboard-path}' | base64 -d
+```
+Uninstall with 
+```bash
+kubectl delete -k manifests/
+```
+
+## Docker
+
+```bash
+docker run -d \
+  -p 5000:5000 \
+  -e CANARY_TOKEN_URL="http://your-canary-token-url" \
+  --name krawl \
+  ghcr.io/blessedrebus/krawl:latest
+```
+
+## Docker Compose
+
+```bash
+docker-compose up -d
+```
+
+## Python 3.11+
+
+Clone the repository
+
+```bash
+git clone https://github.com/blessedrebus/krawl.git
+cd krawl/src
+```
+Run the server
+```bash
+python3 server.py
+```
+
+Visit
+
+`http://localhost:5000`
+
+To access the dashboard
+
+`http://localhost:5000/dashboard-secret-path`
+
+## Configuration via Environment Variables
+
+To customize the deception server installation several **environment variables** can be specified.
+
+| Variable | Description | Default |
+|----------|-------------|---------|
+| `PORT` | Server listening port | `5000` |
+| `DELAY` | Response delay in milliseconds | `100` |
+| `LINKS_MIN_LENGTH` | Minimum random link length | `5` |
+| `LINKS_MAX_LENGTH` | Maximum random link length | `15` |
+| `LINKS_MIN_PER_PAGE` | Minimum links per page | `10` |
+| `LINKS_MAX_PER_PAGE` | Maximum links per page | `15` |
+| `MAX_COUNTER` | Initial counter value | `10` |
+| `CANARY_TOKEN_TRIES` | Requests before showing canary token | `10` |
+| `CANARY_TOKEN_URL` | External canary token URL | None |
+| `DASHBOARD_SECRET_PATH` | Custom dashboard path | Auto-generated |
+| `PROBABILITY_ERROR_CODES` | Error response probability (0-100%) | `0` |
+
+## robots.txt
+The actual (juicy) robots.txt configuration is the following
+
+```txt
+Disallow: /admin/
+Disallow: /api/
+Disallow: /backup/
+Disallow: /config/
+Disallow: /database/
+Disallow: /private/
+Disallow: /uploads/
+Disallow: /wp-admin/
+Disallow: /phpMyAdmin/
+Disallow: /admin/login.php
+Disallow: /api/v1/users
+Disallow: /api/v2/secrets
+Disallow: /.env
+Disallow: /credentials.txt
+Disallow: /passwords.txt
+Disallow: /.git/
+Disallow: /backup.sql
+Disallow: /db_backup.sql
+```
+## Wordlists Customization
+
+Edit `wordlists.json` to customize fake data:
+
+```json
+{
+  "usernames": {
+    "prefixes": ["admin", "root", "user"],
+    "suffixes": ["_prod", "_dev", "123"]
+  },
+  "passwords": {
+    "prefixes": ["P@ssw0rd", "Admin"],
+    "simple": ["test", "password"]
+  },
+  "directory_listing": {
+    "files": ["credentials.txt", "backup.sql"],
+    "directories": ["admin/", "backup/"]
+  }
+}
+```
+
+or **values.yaml** in the case of helm chart installation
+
+## Dashboard
+
+Access the dashboard at `http://<server-ip>:<port>/<dashboard-path>`
+
+The attackers' triggered honeypot path and the suspicious activity (such as failed login attempts) are logged
+
+![asd](img/dashboard-1.png)
+
+The top IP Addresses is shown along with top paths and User Agents
+
+![asd](img/dashboard-2.png)
+
+The dashboard shows:
+- Total and unique accesses
+- Suspicious activity detection
+- Honeypot triggers
+- Top IPs, paths, and user-agents
+- Real-time monitoring
+
+### Retrieving Dashboard Path
+
+Check server startup logs
+
+**Python/Docker:**
+```bash
+docker logs krawl | grep "Dashboard available"
+```
+
+**Kubernetes:**
+```bash
+kubectl get secret krawl-server -n krawl-system \
+  -o jsonpath='{.data.dashboard-path}' | base64 -d && echo
+```
+
+**Helm:**
+```bash
+kubectl get secret krawl -n krawl-system \
+  -o jsonpath='{.data.dashboard-path}' | base64 -d && echo
+```
+
+## Deception Techniques
+
+### 1. Robots.txt Honeypots
+Advertises forbidden paths that legitimate crawlers avoid but scanners investigate:
+- `/admin/`, `/backup/`, `/config/`
+- `/credentials.txt`, `/.env`, `/passwords.txt`
+
+### 2. Fake Services
+Mimics real applications:
+- WordPress (`/wp-admin`, `/wp-login.php`)
+- phpMyAdmin (`/phpmyadmin`)
+- Admin panels (`/admin`, `/login`)
+
+### 3. Credential Traps
+Generates realistic but fake:
+- Usernames and passwords
+- API keys and tokens
+- Database connection strings
+- AWS credentials
+
+### 4. Spider Traps
+Infinite random links to waste automated scanner time
+
+### 5. Error Simulation
+Random HTTP errors to appear more realistic
+
+
+### Custom Canary Token
+
+Generate a canary token at [canarytokens.org](https://canarytokens.org) and configure:
+
+```bash
+export CANARY_TOKEN_URL="http://canarytokens.com/..."
+python3 src/server.py
+```
+
+## Contributing
+
+Contributions welcome! Please:
+1. Fork the repository
+2. Create a feature branch
+3. Make your changes
+4. Submit a pull request
+
+
+<div align="center">
+
+## Disclaimer
+
+**This is a deception/honeypot system.**  
+Deploy in isolated environments and monitor carefully for security events.  
+Use responsibly and in compliance with applicable laws and regulations.

deployment.yaml

@@ -0,0 +1,60 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: krawl-server
+  namespace: krawl
+  labels:
+    app: krawl-server
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      app: krawl-server
+  template:
+    metadata:
+      labels:
+        app: krawl-server
+    spec:
+      containers:
+      - name: krawl
+        image: ghcr.io/blessedrebus/krawl:latest
+        imagePullPolicy: Always
+        ports:
+        - containerPort: 5000
+          name: http
+          protocol: TCP
+        envFrom:
+        - configMapRef:
+            name: krawl-config
+        volumeMounts:
+        - name: wordlists
+          mountPath: /app/wordlists.json
+          subPath: wordlists.json
+          readOnly: true
+        resources:
+          requests:
+            memory: "64Mi"
+            cpu: "100m"
+          limits:
+            memory: "256Mi"
+            cpu: "500m"
+        livenessProbe:
+          httpGet:
+            path: /
+            port: 5000
+          initialDelaySeconds: 10
+          periodSeconds: 30
+          timeoutSeconds: 5
+          failureThreshold: 3
+        readinessProbe:
+          httpGet:
+            path: /
+            port: 5000
+          initialDelaySeconds: 5
+          periodSeconds: 10
+          timeoutSeconds: 5
+          failureThreshold: 3
+      volumes:
+      - name: wordlists
+        configMap:
+          name: krawl-wordlists

docker-compose.yaml

@@ -0,0 +1,33 @@
+version: '3.8'
+
+services:
+  krawl:
+    build:
+      context: .
+      dockerfile: Dockerfile
+    container_name: krawl-server
+    ports:
+      - "5000:5000"
+    volumes:
+      - ./wordlists.json:/app/wordlists.json:ro
+    environment:
+      - PORT=5000
+      - DELAY=100
+      - LINKS_MIN_LENGTH=5
+      - LINKS_MAX_LENGTH=15
+      - LINKS_MIN_PER_PAGE=10
+      - LINKS_MAX_PER_PAGE=15
+      - MAX_COUNTER=10
+      - CANARY_TOKEN_TRIES=10
+      - PROBABILITY_ERROR_CODES=0
+      # Optional: Set your canary token URL
+      # - CANARY_TOKEN_URL=http://canarytokens.com/api/users/YOUR_TOKEN/passwords.txt
+      # Optional: Set custom dashboard path (auto-generated if not set)
+      # - DASHBOARD_SECRET_PATH=/my-secret-dashboard
+    restart: unless-stopped
+    healthcheck:
+      test: ["CMD", "python3", "-c", "import requests; requests.get('http://localhost:5000')"]
+      interval: 30s
+      timeout: 5s
+      retries: 3
+      start_period: 10s

helm/.helmignore

@@ -0,0 +1 @@
+.helmignore

helm/Chart.yaml

@@ -0,0 +1,15 @@
+apiVersion: v2
+name: krawl-chart
+description: A Helm chart for Krawl honeypot server
+type: application
+version: 0.1.2
+appVersion: "1.0.0"
+keywords:
+  - honeypot
+  - security
+  - krawl
+maintainers:
+  - name: blessedrebus
+home: https://github.com/blessedrebus/krawl
+sources:
+  - https://github.com/blessedrebus/krawl

helm/NOTES.txt

@@ -0,0 +1,60 @@
+▄▄▄   ▄▄▄ ▄▄▄▄▄▄▄     ▄▄▄▄   ▄▄▄▄  ▄▄▄  ▄▄▄▄ ▄▄▄      
+███ ▄███▀ ███▀▀███▄ ▄██▀▀██▄ ▀███  ███  ███▀ ███      
+███████   ███▄▄███▀ ███  ███  ███  ███  ███  ███      
+███▀███▄  ███▀▀██▄  ███▀▀███  ███▄▄███▄▄███  ███      
+███  ▀███ ███  ▀███ ███  ███   ▀████▀████▀   ████████ 
+              |                                                 
+              | 
+              | 
+              | 
+              | 
+           ||   ||
+           \\(_)//
+          //(___)\\
+          ||     ||
+
+WARNING: This is a krawl/honeypot service. Monitor access logs for security events.
+
+For more information, visit: https://github.com/blessedrebus/krawl
+
+Your krawl honeypot server has been deployed successfully.
+
+{{- if .Values.service.type }}
+
+Service Type: {{ .Values.service.type }}
+{{- if eq .Values.service.type "LoadBalancer" }}
+
+To get the LoadBalancer IP address, run:
+  kubectl get svc {{ include "krawl.fullname" . }} -n {{ .Release.Namespace }}
+
+Once the EXTERNAL-IP is assigned, access your krawl server at:
+  http://<EXTERNAL-IP>:{{ .Values.service.port }}
+{{- else if eq .Values.service.type "NodePort" }}
+
+To get the NodePort, run:
+  export NODE_PORT=$(kubectl get svc {{ include "krawl.fullname" . }} -n {{ .Release.Namespace }} -o jsonpath='{.spec.ports[0].nodePort}')
+  export NODE_IP=$(kubectl get nodes -o jsonpath='{.items[0].status.addresses[0].address}')
+  echo "Access at: http://$NODE_IP:$NODE_PORT"
+{{- else if eq .Values.service.type "ClusterIP" }}
+
+To access the service from your local machine:
+  kubectl port-forward svc/{{ include "krawl.fullname" . }} {{ .Values.service.port }}:{{ .Values.service.port }} -n {{ .Release.Namespace }}
+
+Then access at: http://localhost:{{ .Values.service.port }}
+{{- end }}
+{{- end }}
+
+Dashboard Access:
+  To retrieve the dashboard path, run:
+    kubectl get secret {{ include "krawl.fullname" . }} -n {{ .Release.Namespace }} -o jsonpath='{.data.dashboard-path}' | base64 -d && echo
+  
+  Then access the dashboard at:
+    http://<EXTERNAL-IP>:{{ .Values.service.port }}/<dashboard-path>
+
+{{- if .Values.ingress.enabled }}
+
+Ingress is ENABLED. Your service will be available at:
+{{- range .Values.ingress.hosts }}
+  - http{{ if $.Values.ingress.tls }}s{{ end }}://{{ .host }}
+{{- end }}
+{{- end }}

helm/templates/NOTES.txt

@@ -0,0 +1,60 @@
+▄▄▄   ▄▄▄ ▄▄▄▄▄▄▄     ▄▄▄▄   ▄▄▄▄  ▄▄▄  ▄▄▄▄ ▄▄▄      
+███ ▄███▀ ███▀▀███▄ ▄██▀▀██▄ ▀███  ███  ███▀ ███      
+███████   ███▄▄███▀ ███  ███  ███  ███  ███  ███      
+███▀███▄  ███▀▀██▄  ███▀▀███  ███▄▄███▄▄███  ███      
+███  ▀███ ███  ▀███ ███  ███   ▀████▀████▀   ████████ 
+              |                                                 
+              | 
+              | 
+              | 
+              | 
+           ||   ||
+           \\(_)//
+          //(___)\\
+          ||     ||
+
+WARNING: This is a deception/honeypot service. Monitor access logs for security events.
+
+For more information, visit: https://github.com/blessedrebus/deception
+
+Your deception honeypot server has been deployed successfully.
+
+{{- if .Values.service.type }}
+
+Service Type: {{ .Values.service.type }}
+{{- if eq .Values.service.type "LoadBalancer" }}
+
+To get the LoadBalancer IP address, run:
+  kubectl get svc {{ include "krawl.fullname" . }} -n {{ .Release.Namespace }}
+
+Once the EXTERNAL-IP is assigned, access your deception server at:
+  http://<EXTERNAL-IP>:{{ .Values.service.port }}
+{{- else if eq .Values.service.type "NodePort" }}
+
+To get the NodePort, run:
+  export NODE_PORT=$(kubectl get svc {{ include "krawl.fullname" . }} -n {{ .Release.Namespace }} -o jsonpath='{.spec.ports[0].nodePort}')
+  export NODE_IP=$(kubectl get nodes -o jsonpath='{.items[0].status.addresses[0].address}')
+  echo "Access at: http://$NODE_IP:$NODE_PORT"
+{{- else if eq .Values.service.type "ClusterIP" }}
+
+To access the service from your local machine:
+  kubectl port-forward svc/{{ include "krawl.fullname" . }} {{ .Values.service.port }}:{{ .Values.service.port }} -n {{ .Release.Namespace }}
+
+Then access at: http://localhost:{{ .Values.service.port }}
+{{- end }}
+{{- end }}
+
+Dashboard Access:
+  To retrieve the dashboard path, run:
+    kubectl get secret {{ include "krawl.fullname" . }} -n {{ .Release.Namespace }} -o jsonpath='{.data.dashboard-path}' | base64 -d && echo
+  
+  Then access the dashboard at:
+    http://<EXTERNAL-IP>:{{ .Values.service.port }}/<dashboard-path>
+
+{{- if .Values.ingress.enabled }}
+
+Ingress is ENABLED. Your service will be available at:
+{{- range .Values.ingress.hosts }}
+  - http{{ if $.Values.ingress.tls }}s{{ end }}://{{ .host }}
+{{- end }}
+{{- end }}

helm/templates/_helpers.tpl

@@ -0,0 +1,60 @@
+{{/*
+Expand the name of the chart.
+*/}}
+{{- define "krawl.name" -}}
+{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }}
+{{- end }}
+
+{{/*
+Create a default fully qualified app name.
+*/}}
+{{- define "krawl.fullname" -}}
+{{- if .Values.fullnameOverride }}
+{{- .Values.fullnameOverride | trunc 63 | trimSuffix "-" }}
+{{- else }}
+{{- $name := default .Chart.Name .Values.nameOverride }}
+{{- if contains $name .Release.Name }}
+{{- .Release.Name | trunc 63 | trimSuffix "-" }}
+{{- else }}
+{{- printf "%s-%s" .Release.Name $name | trunc 63 | trimSuffix "-" }}
+{{- end }}
+{{- end }}
+{{- end }}
+
+{{/*
+Create chart name and version as used by the chart label.
+*/}}
+{{- define "krawl.chart" -}}
+{{- printf "%s-%s" .Chart.Name .Chart.Version | replace "+" "_" | trunc 63 | trimSuffix "-" }}
+{{- end }}
+
+{{/*
+Common labels
+*/}}
+{{- define "krawl.labels" -}}
+helm.sh/chart: {{ include "krawl.chart" . }}
+{{ include "krawl.selectorLabels" . }}
+{{- if .Chart.AppVersion }}
+app.kubernetes.io/version: {{ .Chart.AppVersion | quote }}
+{{- end }}
+app.kubernetes.io/managed-by: {{ .Release.Service }}
+{{- end }}
+
+{{/*
+Selector labels
+*/}}
+{{- define "krawl.selectorLabels" -}}
+app.kubernetes.io/name: {{ include "krawl.name" . }}
+app.kubernetes.io/instance: {{ .Release.Name }}
+{{- end }}
+
+{{/*
+Create the name of the service account to use
+*/}}
+{{- define "krawl.serviceAccountName" -}}
+{{- if .Values.serviceAccount.create }}
+{{- default (include "krawl.fullname" .) .Values.serviceAccount.name }}
+{{- else }}
+{{- default "default" .Values.serviceAccount.name }}
+{{- end }}
+{{- end }}

helm/templates/configmap.yaml

@@ -0,0 +1,17 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: {{ include "krawl.fullname" . }}-config
+  labels:
+    {{- include "krawl.labels" . | nindent 4 }}
+data:
+  PORT: {{ .Values.config.port | quote }}
+  DELAY: {{ .Values.config.delay | quote }}
+  LINKS_MIN_LENGTH: {{ .Values.config.linksMinLength | quote }}
+  LINKS_MAX_LENGTH: {{ .Values.config.linksMaxLength | quote }}
+  LINKS_MIN_PER_PAGE: {{ .Values.config.linksMinPerPage | quote }}
+  LINKS_MAX_PER_PAGE: {{ .Values.config.linksMaxPerPage | quote }}
+  MAX_COUNTER: {{ .Values.config.maxCounter | quote }}
+  CANARY_TOKEN_TRIES: {{ .Values.config.canaryTokenTries | quote }}
+  PROBABILITY_ERROR_CODES: {{ .Values.config.probabilityErrorCodes | quote }}
+  CANARY_TOKEN_URL: {{ .Values.config.canaryTokenUrl | quote }}

helm/templates/deployment.yaml

@@ -0,0 +1,84 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: {{ include "krawl.fullname" . }}
+  labels:
+    {{- include "krawl.labels" . | nindent 4 }}
+spec:
+  {{- if not .Values.autoscaling.enabled }}
+  replicas: {{ .Values.replicaCount }}
+  {{- end }}
+  selector:
+    matchLabels:
+      {{- include "krawl.selectorLabels" . | nindent 6 }}
+  template:
+    metadata:
+      {{- with .Values.podAnnotations }}
+      annotations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      labels:
+        {{- include "krawl.selectorLabels" . | nindent 8 }}
+    spec:
+      {{- with .Values.imagePullSecrets }}
+      imagePullSecrets:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.podSecurityContext }}
+      securityContext:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      containers:
+      - name: {{ .Chart.Name }}
+        {{- with .Values.securityContext }}
+        securityContext:
+          {{- toYaml . | nindent 12 }}
+        {{- end }}
+        image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default .Chart.AppVersion }}"
+        imagePullPolicy: {{ .Values.image.pullPolicy }}
+        ports:
+        - name: http
+          containerPort: {{ .Values.config.port }}
+          protocol: TCP
+        envFrom:
+        - configMapRef:
+            name: {{ include "krawl.fullname" . }}-config
+        env:
+        - name: DASHBOARD_SECRET_PATH
+          valueFrom:
+            secretKeyRef:
+              name: {{ include "krawl.fullname" . }}
+              key: dashboard-path
+        volumeMounts:
+        - name: wordlists
+          mountPath: /app/wordlists.json
+          subPath: wordlists.json
+          readOnly: true
+        {{- with .Values.livenessProbe }}
+        livenessProbe:
+          {{- toYaml . | nindent 12 }}
+        {{- end }}
+        {{- with .Values.readinessProbe }}
+        readinessProbe:
+          {{- toYaml . | nindent 12 }}
+        {{- end }}
+        {{- with .Values.resources }}
+        resources:
+          {{- toYaml . | nindent 12 }}
+        {{- end }}
+      volumes:
+      - name: wordlists
+        configMap:
+          name: {{ include "krawl.fullname" . }}-wordlists
+      {{- with .Values.nodeSelector }}
+      nodeSelector:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.affinity }}
+      affinity:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.tolerations }}
+      tolerations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}

helm/templates/hpa.yaml

@@ -0,0 +1,32 @@
+{{- if .Values.autoscaling.enabled }}
+apiVersion: autoscaling/v2
+kind: HorizontalPodAutoscaler
+metadata:
+  name: {{ include "krawl.fullname" . }}
+  labels:
+    {{- include "krawl.labels" . | nindent 4 }}
+spec:
+  scaleTargetRef:
+    apiVersion: apps/v1
+    kind: Deployment
+    name: {{ include "krawl.fullname" . }}
+  minReplicas: {{ .Values.autoscaling.minReplicas }}
+  maxReplicas: {{ .Values.autoscaling.maxReplicas }}
+  metrics:
+  {{- if .Values.autoscaling.targetCPUUtilizationPercentage }}
+  - type: Resource
+    resource:
+      name: cpu
+      target:
+        type: Utilization
+        averageUtilization: {{ .Values.autoscaling.targetCPUUtilizationPercentage }}
+  {{- end }}
+  {{- if .Values.autoscaling.targetMemoryUtilizationPercentage }}
+  - type: Resource
+    resource:
+      name: memory
+      target:
+        type: Utilization
+        averageUtilization: {{ .Values.autoscaling.targetMemoryUtilizationPercentage }}
+  {{- end }}
+{{- end }}

helm/templates/ingress.yaml

@@ -0,0 +1,42 @@
+{{- if .Values.ingress.enabled -}}
+apiVersion: networking.k8s.io/v1
+kind: Ingress
+metadata:
+  name: {{ include "krawl.fullname" . }}
+  namespace: {{ .Release.Namespace }}
+  labels:
+    {{- include "krawl.labels" . | nindent 4 }}
+  {{- with .Values.ingress.annotations }}
+  annotations:
+    {{- toYaml . | nindent 4 }}
+  {{- end }}
+spec:
+  {{- if .Values.ingress.className }}
+  ingressClassName: {{ .Values.ingress.className }}
+  {{- end }}
+  {{- if .Values.ingress.tls }}
+  tls:
+    {{- range .Values.ingress.tls }}
+    - hosts:
+        {{- range .hosts }}
+        - {{ . | quote }}
+        {{- end }}
+      secretName: {{ .secretName }}
+    {{- end }}
+  {{- end }}
+  rules:
+    {{- range .Values.ingress.hosts }}
+    - host: {{ .host | quote }}
+      http:
+        paths:
+          {{- range .paths }}
+          - path: {{ .path }}
+            pathType: {{ .pathType }}
+            backend:
+              service:
+                name: {{ include "krawl.fullname" $ }}
+                port:
+                  number: {{ $.Values.service.port }}
+          {{- end }}
+    {{- end }}
+{{- end }}

helm/templates/network-policy.yaml

@@ -0,0 +1,24 @@
+{{- if .Values.networkPolicy.enabled -}}
+apiVersion: networking.k8s.io/v1
+kind: NetworkPolicy
+metadata:
+  name: {{ include "krawl.fullname" . }}
+  labels:
+    {{- include "krawl.labels" . | nindent 4 }}
+spec:
+  podSelector:
+    matchLabels:
+      {{- include "krawl.selectorLabels" . | nindent 6 }}
+  {{- with .Values.networkPolicy.policyTypes }}
+  policyTypes:
+    {{- toYaml . | nindent 4 }}
+  {{- end }}
+  {{- with .Values.networkPolicy.ingress }}
+  ingress:
+    {{- toYaml . | nindent 4 }}
+  {{- end }}
+  {{- with .Values.networkPolicy.egress }}
+  egress:
+    {{- toYaml . | nindent 4 }}
+  {{- end }}
+{{- end }}

helm/templates/secret.yaml

@@ -0,0 +1,16 @@
+{{- $secret := (lookup "v1" "Secret" .Release.Namespace (include "krawl.fullname" .)) -}}
+{{- $dashboardPath := "" -}}
+{{- if and $secret $secret.data -}}
+  {{- $dashboardPath = index $secret.data "dashboard-path" | b64dec -}}
+{{- else -}}
+  {{- $dashboardPath = printf "/%s" (randAlphaNum 32) -}}
+{{- end -}}
+apiVersion: v1
+kind: Secret
+metadata:
+  name: {{ include "krawl.fullname" . }}
+  labels:
+    {{- include "krawl.labels" . | nindent 4 }}
+type: Opaque
+stringData:
+  dashboard-path: {{ $dashboardPath | quote }}

helm/templates/service.yaml

@@ -0,0 +1,26 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: {{ include "krawl.fullname" . }}
+  labels:
+    {{- include "krawl.labels" . | nindent 4 }}
+  {{- with .Values.service.annotations }}
+  annotations:
+    {{- toYaml . | nindent 4 }}
+  {{- end }}
+spec:
+  type: {{ .Values.service.type }}
+  {{- if .Values.service.externalTrafficPolicy }}
+  externalTrafficPolicy: {{ .Values.service.externalTrafficPolicy }}
+  {{- end }}
+  sessionAffinity: ClientIP
+  sessionAffinityConfig:
+    clientIP:
+      timeoutSeconds: 10800
+  ports:
+  - port: {{ .Values.service.port }}
+    targetPort: http
+    protocol: TCP
+    name: http
+  selector:
+    {{- include "krawl.selectorLabels" . | nindent 4 }}

helm/templates/wordlists-configmap.yaml

@@ -0,0 +1,9 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: {{ include "krawl.fullname" . }}-wordlists
+  labels:
+    {{- include "krawl.labels" . | nindent 4 }}
+data:
+  wordlists.json: |
+    {{- .Values.wordlists | toJson | nindent 4 }}

helm/values.yaml

@@ -0,0 +1,295 @@
+replicaCount: 1
+
+image:
+  repository: ghcr.io/blessedrebus/krawl
+  pullPolicy: Always
+  tag: "latest"
+
+imagePullSecrets: []
+nameOverride: "krawl"
+fullnameOverride: ""
+
+serviceAccount:
+  create: false
+  annotations: {}
+  name: ""
+
+podAnnotations: {}
+
+podSecurityContext: {}
+
+securityContext: {}
+
+service:
+  type: LoadBalancer
+  port: 5000
+  annotations: {}
+  # Preserve source IP when using LoadBalancer
+  externalTrafficPolicy: Local
+
+ingress:
+  enabled: true
+  className: "nginx"
+  annotations: {}
+  hosts:
+    - host: krawl.example.com
+      paths:
+        - path: /
+          pathType: Prefix
+  tls: []
+  #  - secretName: krawl-tls
+  #    hosts:
+  #      - krawl.example.com
+
+resources:
+  limits:
+    cpu: 500m
+    memory: 256Mi
+  requests:
+    cpu: 100m
+    memory: 64Mi
+
+livenessProbe:
+  httpGet:
+    path: /
+    port: http
+  initialDelaySeconds: 10
+  periodSeconds: 30
+  timeoutSeconds: 5
+  failureThreshold: 3
+
+readinessProbe:
+  httpGet:
+    path: /
+    port: http
+  initialDelaySeconds: 5
+  periodSeconds: 10
+  timeoutSeconds: 5
+  failureThreshold: 3
+
+autoscaling:
+  enabled: false
+  minReplicas: 1
+  maxReplicas: 5
+  targetCPUUtilizationPercentage: 70
+  targetMemoryUtilizationPercentage: 80
+
+nodeSelector: {}
+
+tolerations: []
+
+affinity: {}
+
+# Application configuration
+config:
+  port: 5000
+  delay: 100
+  linksMinLength: 5
+  linksMaxLength: 15
+  linksMinPerPage: 10
+  linksMaxPerPage: 15
+  maxCounter: 10
+  canaryTokenTries: 10
+  probabilityErrorCodes: 0
+#  canaryTokenUrl: set-your-canary-token-url-here
+
+networkPolicy:
+  enabled: true
+  policyTypes:
+    - Ingress
+    - Egress
+  ingress:
+    - from:
+      - podSelector: {}
+      - namespaceSelector: {}
+      - ipBlock:
+          cidr: 0.0.0.0/0
+      ports:
+      - protocol: TCP
+        port: 5000
+  egress:
+    - to:
+      - namespaceSelector: {}
+      - ipBlock:
+          cidr: 0.0.0.0/0
+      ports:
+      - protocol: TCP
+      - protocol: UDP
+
+# Wordlists configuration
+wordlists:
+  usernames:
+    prefixes:
+      - admin
+      - user
+      - developer
+      - root
+      - system
+      - db
+      - api
+      - service
+      - deploy
+      - test
+      - prod
+      - backup
+      - monitor
+      - jenkins
+      - webapp
+    suffixes:
+      - ""
+      - "_prod"
+      - "_dev"
+      - "_test"
+      - "123"
+      - "2024"
+      - "_backup"
+      - "_admin"
+      - "01"
+      - "02"
+      - "_user"
+      - "_service"
+      - "_api"
+  passwords:
+    prefixes:
+      - P@ssw0rd
+      - Passw0rd
+      - Admin
+      - Secret
+      - Welcome
+      - System
+      - Database
+      - Secure
+      - Master
+      - Root
+    simple:
+      - test
+      - demo
+      - temp
+      - change
+      - password
+      - admin
+      - letmein
+      - welcome
+      - default
+      - sample
+  emails:
+    domains:
+      - example.com
+      - company.com
+      - localhost.com
+      - test.com
+      - domain.com
+      - corporate.com
+      - internal.net
+      - enterprise.com
+      - business.org
+  api_keys:
+    prefixes:
+      - sk_live_
+      - sk_test_
+      - api_
+      - key_
+      - token_
+      - access_
+      - secret_
+      - prod_
+      - ""
+  databases:
+    names:
+      - production
+      - prod_db
+      - main_db
+      - app_database
+      - users_db
+      - customer_data
+      - analytics
+      - staging_db
+      - dev_database
+      - wordpress
+      - ecommerce
+      - crm_db
+      - inventory
+    hosts:
+      - localhost
+      - db.internal
+      - mysql.local
+      - postgres.internal
+      - 127.0.0.1
+      - db-server-01
+      - database.prod
+      - sql.company.com
+  applications:
+    names:
+      - WebApp
+      - API Gateway
+      - Dashboard
+      - Admin Panel
+      - CMS
+      - Portal
+      - Manager
+      - Console
+      - Control Panel
+      - Backend
+  users:
+    roles:
+      - Administrator
+      - Developer
+      - Manager
+      - User
+      - Guest
+      - Moderator
+      - Editor
+      - Viewer
+      - Analyst
+      - Support
+  directory_listing:
+    files:
+      - admin.txt
+      - test.exe
+      - backup.sql
+      - database.sql
+      - db_backup.sql
+      - dump.sql
+      - config.php
+      - credentials.txt
+      - passwords.txt
+      - users.csv
+      - .env
+      - id_rsa
+      - id_rsa.pub
+      - private_key.pem
+      - api_keys.json
+      - secrets.yaml
+      - admin_notes.txt
+      - settings.ini
+      - database.yml
+      - wp-config.php
+      - .htaccess
+      - server.key
+      - cert.pem
+      - shadow.bak
+      - passwd.old
+    directories:
+      - uploads/
+      - backups/
+      - logs/
+      - temp/
+      - cache/
+      - private/
+      - config/
+      - admin/
+      - database/
+      - backup/
+      - old/
+      - archive/
+      - .git/
+      - keys/
+      - credentials/
+  error_codes:
+    - 400
+    - 401
+    - 403
+    - 404
+    - 500
+    - 502
+    - 503

manifests/configmap.yaml

@@ -0,0 +1,16 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: krawl-config
+  namespace: krawl-system
+data:
+  PORT: "5000"
+  DELAY: "100"
+  LINKS_MIN_LENGTH: "5"
+  LINKS_MAX_LENGTH: "15"
+  LINKS_MIN_PER_PAGE: "10"
+  LINKS_MAX_PER_PAGE: "15"
+  MAX_COUNTER: "10"
+  CANARY_TOKEN_TRIES: "10"
+  PROBABILITY_ERROR_CODES: "0"
+#  CANARY_TOKEN_URL: set-your-canary-token-url-here

manifests/deployment.yaml

@@ -0,0 +1,60 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: krawl-server
+  namespace: krawl-system
+  labels:
+    app: krawl-server
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      app: krawl-server
+  template:
+    metadata:
+      labels:
+        app: krawl-server
+    spec:
+      containers:
+      - name: krawl
+        image: ghcr.io/blessedrebus/krawl:latest
+        imagePullPolicy: Always
+        ports:
+        - containerPort: 5000
+          name: http
+          protocol: TCP
+        envFrom:
+        - configMapRef:
+            name: krawl-config
+        volumeMounts:
+        - name: wordlists
+          mountPath: /app/wordlists.json
+          subPath: wordlists.json
+          readOnly: true
+        resources:
+          requests:
+            memory: "64Mi"
+            cpu: "100m"
+          limits:
+            memory: "256Mi"
+            cpu: "500m"
+        livenessProbe:
+          httpGet:
+            path: /
+            port: 5000
+          initialDelaySeconds: 10
+          periodSeconds: 30
+          timeoutSeconds: 5
+          failureThreshold: 3
+        readinessProbe:
+          httpGet:
+            path: /
+            port: 5000
+          initialDelaySeconds: 5
+          periodSeconds: 10
+          timeoutSeconds: 5
+          failureThreshold: 3
+      volumes:
+      - name: wordlists
+        configMap:
+          name: krawl-wordlists

manifests/hpa.yaml

@@ -0,0 +1,26 @@
+# Optional: HorizontalPodAutoscaler for auto-scaling
+apiVersion: autoscaling/v2
+kind: HorizontalPodAutoscaler
+metadata:
+  name: krawl-hpa
+  namespace: krawl-system
+spec:
+  scaleTargetRef:
+    apiVersion: apps/v1
+    kind: Deployment
+    name: krawl-server
+  minReplicas: 1
+  maxReplicas: 5
+  metrics:
+  - type: Resource
+    resource:
+      name: cpu
+      target:
+        type: Utilization
+        averageUtilization: 70
+  - type: Resource
+    resource:
+      name: memory
+      target:
+        type: Utilization
+        averageUtilization: 80

manifests/ingress.yaml

@@ -0,0 +1,24 @@
+apiVersion: networking.k8s.io/v1
+kind: Ingress
+metadata:
+  name: krawl-ingress
+  namespace: krawl-system
+  annotations:
+    nginx.ingress.kubernetes.io/rewrite-target: /
+spec:
+  ingressClassName: nginx
+  rules:
+  - host: krawl.example.com  # Change to your domain
+    http:
+      paths:
+      - path: /
+        pathType: Prefix
+        backend:
+          service:
+            name: krawl-server
+            port:
+              number: 5000
+  # tls:
+  # - hosts:
+  #   - krawl.example.com
+  #   secretName: krawl-tls

manifests/kustomization.yaml

@@ -0,0 +1,14 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+resources:
+  - namespace.yaml
+  - configmap.yaml
+  - wordlists-configmap.yaml
+  - deployment.yaml
+  - service.yaml
+  - network-policy.yaml
+  - ingress.yaml
+  - hpa.yaml
+
+namespace: krawl-system

manifests/namespace.yaml

@@ -0,0 +1,4 @@
+apiVersion: v1
+kind: Namespace
+metadata:
+  name: krawl-system

manifests/network-policy.yaml

@@ -0,0 +1,29 @@
+apiVersion: networking.k8s.io/v1
+kind: NetworkPolicy
+metadata:
+  name: krawl-network-policy
+  namespace: krawl-system
+spec:
+  podSelector:
+    matchLabels:
+      app: krawl-server
+  policyTypes:
+  - Ingress
+  - Egress
+  ingress:
+  - from:
+    - podSelector: {}
+    - namespaceSelector: {}
+    - ipBlock:
+        cidr: 0.0.0.0/0 
+    ports:
+    - protocol: TCP
+      port: 5000
+  egress:
+  - to:
+    - namespaceSelector: {}
+    - ipBlock:
+        cidr: 0.0.0.0/0
+    ports:
+    - protocol: TCP
+    - protocol: UDP

manifests/service.yaml

@@ -0,0 +1,16 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: krawl-server
+  namespace: krawl-system
+  labels:
+    app: krawl-server
+spec:
+  type: LoadBalancer
+  ports:
+  - port: 5000
+    targetPort: 5000
+    protocol: TCP
+    name: http
+  selector:
+    app: krawl-server

manifests/wordlists-configmap.yaml

@@ -0,0 +1,205 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: krawl-wordlists
+  namespace: krawl-system
+data:
+  wordlists.json: |
+    {
+      "usernames": {
+        "prefixes": [
+          "admin",
+          "user",
+          "developer",
+          "root",
+          "system",
+          "db",
+          "api",
+          "service",
+          "deploy",
+          "test",
+          "prod",
+          "backup",
+          "monitor",
+          "jenkins",
+          "webapp"
+        ],
+        "suffixes": [
+          "",
+          "_prod",
+          "_dev",
+          "_test",
+          "123",
+          "2024",
+          "_backup",
+          "_admin",
+          "01",
+          "02",
+          "_user",
+          "_service",
+          "_api"
+        ]
+      },
+      "passwords": {
+        "prefixes": [
+          "P@ssw0rd",
+          "Passw0rd",
+          "Admin",
+          "Secret",
+          "Welcome",
+          "System",
+          "Database",
+          "Secure",
+          "Master",
+          "Root"
+        ],
+        "simple": [
+          "test",
+          "demo",
+          "temp",
+          "change",
+          "password",
+          "admin",
+          "letmein",
+          "welcome",
+          "default",
+          "sample"
+        ]
+      },
+      "emails": {
+        "domains": [
+          "example.com",
+          "company.com",
+          "localhost.com",
+          "test.com",
+          "domain.com",
+          "corporate.com",
+          "internal.net",
+          "enterprise.com",
+          "business.org"
+        ]
+      },
+      "api_keys": {
+        "prefixes": [
+          "sk_live_",
+          "sk_test_",
+          "api_",
+          "key_",
+          "token_",
+          "access_",
+          "secret_",
+          "prod_",
+          ""
+        ]
+      },
+      "databases": {
+        "names": [
+          "production",
+          "prod_db",
+          "main_db",
+          "app_database",
+          "users_db",
+          "customer_data",
+          "analytics",
+          "staging_db",
+          "dev_database",
+          "wordpress",
+          "ecommerce",
+          "crm_db",
+          "inventory"
+        ],
+        "hosts": [
+          "localhost",
+          "db.internal",
+          "mysql.local",
+          "postgres.internal",
+          "127.0.0.1",
+          "db-server-01",
+          "database.prod",
+          "sql.company.com"
+        ]
+      },
+      "applications": {
+        "names": [
+          "WebApp",
+          "API Gateway",
+          "Dashboard",
+          "Admin Panel",
+          "CMS",
+          "Portal",
+          "Manager",
+          "Console",
+          "Control Panel",
+          "Backend"
+        ]
+      },
+      "users": {
+        "roles": [
+          "Administrator",
+          "Developer",
+          "Manager",
+          "User",
+          "Guest",
+          "Moderator",
+          "Editor",
+          "Viewer",
+          "Analyst",
+          "Support"
+        ]
+      },
+      "directory_listing": {
+        "files": [
+          "admin.txt",
+          "test.exe",
+          "backup.sql",
+          "database.sql",
+          "db_backup.sql",
+          "dump.sql",
+          "config.php",
+          "credentials.txt",
+          "passwords.txt",
+          "users.csv",
+          ".env",
+          "id_rsa",
+          "id_rsa.pub",
+          "private_key.pem",
+          "api_keys.json",
+          "secrets.yaml",
+          "admin_notes.txt",
+          "settings.ini",
+          "database.yml",
+          "wp-config.php",
+          ".htaccess",
+          "server.key",
+          "cert.pem",
+          "shadow.bak",
+          "passwd.old"
+        ],
+        "directories": [
+          "uploads/",
+          "backups/",
+          "logs/",
+          "temp/",
+          "cache/",
+          "private/",
+          "config/",
+          "admin/",
+          "database/",
+          "backup/",
+          "old/",
+          "archive/",
+          ".git/",
+          "keys/",
+          "credentials/"
+        ]
+      },
+      "error_codes": [
+        400,
+        401,
+        403,
+        404,
+        500,
+        502,
+        503
+      ]
+    }

src/config.py

@@ -0,0 +1,48 @@
+#!/usr/bin/env python3
+
+import os
+from dataclasses import dataclass
+from typing import Optional, Tuple
+
+
+@dataclass
+class Config:
+    """Configuration class for the deception server"""
+    port: int = 5000
+    delay: int = 100  # milliseconds
+    links_length_range: Tuple[int, int] = (5, 15)
+    links_per_page_range: Tuple[int, int] = (10, 15)
+    char_space: str = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789'
+    max_counter: int = 10
+    canary_token_url: Optional[str] = None
+    canary_token_tries: int = 10
+    dashboard_secret_path: str = None
+    api_server_url: Optional[str] = None
+    api_server_port: int = 8080
+    api_server_path: str = "/api/v2/users"
+    probability_error_codes: int = 0  # Percentage (0-100)
+
+    @classmethod
+    def from_env(cls) -> 'Config':
+        """Create configuration from environment variables"""
+        return cls(
+            port=int(os.getenv('PORT', 5000)),
+            delay=int(os.getenv('DELAY', 100)),
+            links_length_range=(
+                int(os.getenv('LINKS_MIN_LENGTH', 5)),
+                int(os.getenv('LINKS_MAX_LENGTH', 15))
+            ),
+            links_per_page_range=(
+                int(os.getenv('LINKS_MIN_PER_PAGE', 10)),
+                int(os.getenv('LINKS_MAX_PER_PAGE', 15))
+            ),
+            char_space=os.getenv('CHAR_SPACE', 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789'),
+            max_counter=int(os.getenv('MAX_COUNTER', 10)),
+            canary_token_url=os.getenv('CANARY_TOKEN_URL'),
+            canary_token_tries=int(os.getenv('CANARY_TOKEN_TRIES', 10)),
+            dashboard_secret_path=os.getenv('DASHBOARD_SECRET_PATH', f'/{os.urandom(16).hex()}'),
+            api_server_url=os.getenv('API_SERVER_URL'),
+            api_server_port=int(os.getenv('API_SERVER_PORT', 8080)),
+            api_server_path=os.getenv('API_SERVER_PATH', '/api/v2/users'),
+            probability_error_codes=int(os.getenv('PROBABILITY_ERROR_CODES', 5))
+        )

src/dashboard_template.py

@@ -0,0 +1,214 @@
+#!/usr/bin/env python3
+
+"""
+Dashboard template for viewing honeypot statistics.
+Customize this template to change the dashboard appearance.
+"""
+
+
+def generate_dashboard(stats: dict) -> str:
+    """Generate dashboard HTML with access statistics"""
+    
+    top_ips_rows = '\n'.join([
+        f'<tr><td class="rank">{i+1}</td><td>{ip}</td><td>{count}</td></tr>'
+        for i, (ip, count) in enumerate(stats['top_ips'])
+    ]) or '<tr><td colspan="3" style="text-align:center;">No data</td></tr>'
+
+    # Generate paths rows
+    top_paths_rows = '\n'.join([
+        f'<tr><td class="rank">{i+1}</td><td>{path}</td><td>{count}</td></tr>'
+        for i, (path, count) in enumerate(stats['top_paths'])
+    ]) or '<tr><td colspan="3" style="text-align:center;">No data</td></tr>'
+
+    # Generate User-Agent rows
+    top_ua_rows = '\n'.join([
+        f'<tr><td class="rank">{i+1}</td><td style="word-break: break-all;">{ua[:80]}</td><td>{count}</td></tr>'
+        for i, (ua, count) in enumerate(stats['top_user_agents'])
+    ]) or '<tr><td colspan="3" style="text-align:center;">No data</td></tr>'
+
+    # Generate suspicious accesses rows
+    suspicious_rows = '\n'.join([
+        f'<tr><td>{log["ip"]}</td><td>{log["path"]}</td><td style="word-break: break-all;">{log["user_agent"][:60]}</td><td>{log["timestamp"].split("T")[1][:8]}</td></tr>'
+        for log in stats['recent_suspicious'][-10:]
+    ]) or '<tr><td colspan="4" style="text-align:center;">No suspicious activity detected</td></tr>'
+
+    return f"""<!DOCTYPE html>
+<html>
+<head>
+    <meta charset="UTF-8">
+    <title>Krawl Dashboard</title>
+    <style>
+        body {{
+            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+            background-color: #0d1117;
+            color: #c9d1d9;
+            margin: 0;
+            padding: 20px;
+        }}
+        .container {{
+            max-width: 1400px;
+            margin: 0 auto;
+        }}
+        h1 {{
+            color: #58a6ff;
+            text-align: center;
+            margin-bottom: 40px;
+        }}
+        .stats-grid {{
+            display: grid;
+            grid-template-columns: repeat(auto-fit, minmax(200px, 1fr));
+            gap: 20px;
+            margin-bottom: 40px;
+        }}
+        .stat-card {{
+            background: #161b22;
+            border: 1px solid #30363d;
+            border-radius: 6px;
+            padding: 20px;
+            text-align: center;
+        }}
+        .stat-card.alert {{
+            border-color: #f85149;
+        }}
+        .stat-value {{
+            font-size: 36px;
+            font-weight: bold;
+            color: #58a6ff;
+        }}
+        .stat-value.alert {{
+            color: #f85149;
+        }}
+        .stat-label {{
+            font-size: 14px;
+            color: #8b949e;
+            margin-top: 5px;
+        }}
+        .table-container {{
+            background: #161b22;
+            border: 1px solid #30363d;
+            border-radius: 6px;
+            padding: 20px;
+            margin-bottom: 20px;
+        }}
+        h2 {{
+            color: #58a6ff;
+            margin-top: 0;
+        }}
+        table {{
+            width: 100%;
+            border-collapse: collapse;
+        }}
+        th, td {{
+            padding: 12px;
+            text-align: left;
+            border-bottom: 1px solid #30363d;
+        }}
+        th {{
+            background: #0d1117;
+            color: #58a6ff;
+            font-weight: 600;
+        }}
+        tr:hover {{
+            background: #1c2128;
+        }}
+        .rank {{
+            color: #8b949e;
+            font-weight: bold;
+        }}
+        .alert-section {{
+            background: #1c1917;
+            border-left: 4px solid #f85149;
+        }}
+    </style>
+</head>
+<body>
+    <div class="container">
+        <h1>&#128375;&#65039; Krawl Dashboard</h1>
+        
+        <div class="stats-grid">
+            <div class="stat-card">
+                <div class="stat-value">{stats['total_accesses']}</div>
+                <div class="stat-label">Total Accesses</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{stats['unique_ips']}</div>
+                <div class="stat-label">Unique IPs</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{stats['unique_paths']}</div>
+                <div class="stat-label">Unique Paths</div>
+            </div>
+            <div class="stat-card alert">
+                <div class="stat-value alert">{stats['suspicious_accesses']}</div>
+                <div class="stat-label">Suspicious Accesses</div>
+            </div>
+        </div>
+
+        <div class="table-container alert-section">
+            <h2>&#9888;&#65039; Recent Suspicious Activity</h2>
+            <table>
+                <thead>
+                    <tr>
+                        <th>IP Address</th>
+                        <th>Path</th>
+                        <th>User-Agent</th>
+                        <th>Time</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {suspicious_rows}
+                </tbody>
+            </table>
+        </div>
+
+        <div class="table-container">
+            <h2>Top IP Addresses</h2>
+            <table>
+                <thead>
+                    <tr>
+                        <th>#</th>
+                        <th>IP Address</th>
+                        <th>Access Count</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {top_ips_rows}
+                </tbody>
+            </table>
+        </div>
+
+        <div class="table-container">
+            <h2>Top Paths</h2>
+            <table>
+                <thead>
+                    <tr>
+                        <th>#</th>
+                        <th>Path</th>
+                        <th>Access Count</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {top_paths_rows}
+                </tbody>
+            </table>
+        </div>
+
+        <div class="table-container">
+            <h2>Top User-Agents</h2>
+            <table>
+                <thead>
+                    <tr>
+                        <th>#</th>
+                        <th>User-Agent</th>
+                        <th>Count</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {top_ua_rows}
+                </tbody>
+            </table>
+        </div>
+    </div>
+</body>
+</html>
+"""

src/generators.py

@@ -0,0 +1,190 @@
+#!/usr/bin/env python3
+
+"""
+Generators for creating random fake data (credentials, API keys, etc.)
+"""
+
+import random
+import string
+import json
+from templates import html_templates
+from wordlists import get_wordlists
+
+
+def random_username() -> str:
+    """Generate random username"""
+    wl = get_wordlists()
+    return random.choice(wl.username_prefixes) + random.choice(wl.username_suffixes)
+
+
+def random_password() -> str:
+    """Generate random password"""
+    wl = get_wordlists()
+    templates = [
+        lambda: ''.join(random.choices(string.ascii_letters + string.digits, k=12)),
+        lambda: f"{random.choice(wl.password_prefixes)}{random.randint(100, 999)}!",
+        lambda: f"{random.choice(wl.simple_passwords)}{random.randint(1000, 9999)}",
+        lambda: ''.join(random.choices(string.ascii_lowercase, k=8)),
+    ]
+    return random.choice(templates)()
+
+
+def random_email(username: str = None) -> str:
+    """Generate random email"""
+    wl = get_wordlists()
+    if not username:
+        username = random_username()
+    return f"{username}@{random.choice(wl.email_domains)}"
+
+
+def random_api_key() -> str:
+    """Generate random API key"""
+    wl = get_wordlists()
+    key = ''.join(random.choices(string.ascii_letters + string.digits, k=32))
+    return random.choice(wl.api_key_prefixes) + key
+
+
+def random_database_name() -> str:
+    """Generate random database name"""
+    wl = get_wordlists()
+    return random.choice(wl.database_names)
+
+
+def credentials_txt() -> str:
+    """Generate fake credentials.txt with random data"""
+    content = "# Production Credentials\n\n"
+    for i in range(random.randint(3, 7)):
+        username = random_username()
+        password = random_password()
+        content += f"{username}:{password}\n"
+    return content
+
+
+def passwords_txt() -> str:
+    """Generate fake passwords.txt with random data"""
+    content = "# Password List\n"
+    content += f"Admin Password: {random_password()}\n"
+    content += f"Database Password: {random_password()}\n"
+    content += f"API Key: {random_api_key()}\n\n"
+    content += "User Passwords:\n"
+    for i in range(random.randint(5, 10)):
+        username = random_username()
+        password = random_password()
+        content += f"{username} = {password}\n"
+    return content
+
+
+def users_json() -> str:
+    """Generate fake users.json with random data"""
+    wl = get_wordlists()
+    users = []
+    for i in range(random.randint(3, 8)):
+        username = random_username()
+        users.append({
+            "id": i + 1,
+            "username": username,
+            "email": random_email(username),
+            "password": random_password(),
+            "role": random.choice(wl.user_roles),
+            "api_token": random_api_key()
+        })
+    return json.dumps({"users": users}, indent=2)
+
+
+def api_keys_json() -> str:
+    """Generate fake api_keys.json with random data"""
+    keys = {
+        "stripe": {
+            "public_key": "pk_live_" + ''.join(random.choices(string.ascii_letters + string.digits, k=24)),
+            "secret_key": random_api_key()
+        },
+        "aws": {
+            "access_key_id": "AKIA" + ''.join(random.choices(string.ascii_uppercase + string.digits, k=16)),
+            "secret_access_key": ''.join(random.choices(string.ascii_letters + string.digits + '+/', k=40))
+        },
+        "sendgrid": {
+            "api_key": "SG." + ''.join(random.choices(string.ascii_letters + string.digits, k=48))
+        },
+        "twilio": {
+            "account_sid": "AC" + ''.join(random.choices(string.ascii_lowercase + string.digits, k=32)),
+            "auth_token": ''.join(random.choices(string.ascii_lowercase + string.digits, k=32))
+        }
+    }
+    return json.dumps(keys, indent=2)
+
+
+def api_response(path: str) -> str:
+    """Generate fake API JSON responses with random data"""
+    wl = get_wordlists()
+    
+    def random_users(count: int = 3):
+        users = []
+        for i in range(count):
+            username = random_username()
+            users.append({
+                "id": i + 1,
+                "username": username,
+                "email": random_email(username),
+                "role": random.choice(wl.user_roles)
+            })
+        return users
+    
+    responses = {
+        '/api/users': json.dumps({
+            "users": random_users(random.randint(2, 5)),
+            "total": random.randint(50, 500)
+        }, indent=2),
+        '/api/v1/users': json.dumps({
+            "status": "success",
+            "data": [{
+                "id": random.randint(1, 100),
+                "name": random_username(),
+                "api_key": random_api_key()
+            }]
+        }, indent=2),
+        '/api/v2/secrets': json.dumps({
+            "database": {
+                "host": random.choice(wl.database_hosts),
+                "username": random_username(),
+                "password": random_password(),
+                "database": random_database_name()
+            },
+            "api_keys": {
+                "stripe": random_api_key(),
+                "aws": 'AKIA' + ''.join(random.choices(string.ascii_uppercase + string.digits, k=16))
+            }
+        }, indent=2),
+        '/api/config': json.dumps({
+            "app_name": random.choice(wl.application_names),
+            "debug": random.choice([True, False]),
+            "secret_key": random_api_key(),
+            "database_url": f"postgresql://{random_username()}:{random_password()}@localhost/{random_database_name()}"
+        }, indent=2),
+        '/.env': f"""APP_NAME={random.choice(wl.application_names)}
+DEBUG={random.choice(['true', 'false'])}
+APP_KEY=base64:{''.join(random.choices(string.ascii_letters + string.digits, k=32))}=
+DB_CONNECTION=mysql
+DB_HOST=127.0.0.1
+DB_PORT=3306
+DB_DATABASE={random_database_name()}
+DB_USERNAME={random_username()}
+DB_PASSWORD={random_password()}
+AWS_ACCESS_KEY_ID=AKIA{''.join(random.choices(string.ascii_uppercase + string.digits, k=16))}
+AWS_SECRET_ACCESS_KEY={''.join(random.choices(string.ascii_letters + string.digits + '+/', k=40))}
+STRIPE_SECRET={random_api_key()}
+"""
+    }
+    return responses.get(path, json.dumps({"error": "Not found"}, indent=2))
+
+
+def directory_listing(path: str) -> str:
+    """Generate fake directory listing using wordlists"""
+    wl = get_wordlists()
+    
+    files = wl.directory_files
+    dirs = wl.directory_dirs
+    
+    selected_files = [(f, random.randint(1024, 1024*1024)) 
+                      for f in random.sample(files, min(6, len(files)))]
+    
+    return html_templates.directory_listing(path, dirs, selected_files)

src/handler.py

@@ -0,0 +1,342 @@
+#!/usr/bin/env python3
+
+import random
+import time
+from http.server import BaseHTTPRequestHandler
+from typing import Optional, List
+from datetime import datetime
+
+from config import Config
+from tracker import AccessTracker
+from templates import html_templates
+from templates.dashboard_template import generate_dashboard
+from generators import (
+    credentials_txt, passwords_txt, users_json, api_keys_json,
+    api_response, directory_listing
+)
+from wordlists import get_wordlists
+
+
+class Handler(BaseHTTPRequestHandler):
+    """HTTP request handler for the deception server"""
+    webpages: Optional[List[str]] = None
+    config: Config = None
+    tracker: AccessTracker = None
+    counter: int = 0
+
+    def _get_client_ip(self) -> str:
+        """Extract client IP address from request, checking proxy headers first"""
+        # Headers might not be available during early error logging
+        if hasattr(self, 'headers') and self.headers:
+            # Check X-Forwarded-For header (set by load balancers/proxies)
+            forwarded_for = self.headers.get('X-Forwarded-For')
+            if forwarded_for:
+                # X-Forwarded-For can contain multiple IPs, get the first (original client)
+                return forwarded_for.split(',')[0].strip()
+            
+            # Check X-Real-IP header (set by nginx and other proxies)
+            real_ip = self.headers.get('X-Real-IP')
+            if real_ip:
+                return real_ip.strip()
+        
+        # Fallback to direct connection IP
+        return self.client_address[0]
+
+    def _get_user_agent(self) -> str:
+        """Extract user agent from request"""
+        return self.headers.get('User-Agent', '')
+
+    def _should_return_error(self) -> bool:
+        """Check if we should return an error based on probability"""
+        if self.config.probability_error_codes <= 0:
+            return False
+        return random.randint(1, 100) <= self.config.probability_error_codes
+
+    def _get_random_error_code(self) -> int:
+        """Get a random error code from wordlists"""
+        wl = get_wordlists()
+        error_codes = wl.error_codes
+        if not error_codes:
+            error_codes = [400, 401, 403, 404, 500, 502, 503]
+        return random.choice(error_codes)
+
+    def generate_page(self, seed: str) -> str:
+        """Generate a webpage containing random links or canary token"""
+        random.seed(seed)
+        num_pages = random.randint(*self.config.links_per_page_range)
+
+        html = f"""<!DOCTYPE html>
+<html>
+<head>
+    <meta charset="UTF-8">
+    <title>Krawl</title>
+    <style>
+        body {{
+            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+            background-color: #0d1117;
+            color: #c9d1d9;
+            margin: 0;
+            padding: 40px 20px;
+            min-height: 100vh;
+            display: flex;
+            flex-direction: column;
+            align-items: center;
+        }}
+        .container {{
+            max-width: 1200px;
+            width: 100%;
+        }}
+        h1 {{
+            color: #f85149;
+            text-align: center;
+            font-size: 48px;
+            margin: 60px 0 30px;
+        }}
+        .counter {{
+            color: #f85149;
+            text-align: center;
+            font-size: 56px;
+            font-weight: bold;
+            margin-bottom: 60px;
+        }}
+        .links-container {{
+            display: flex;
+            flex-direction: column;
+            gap: 20px;
+            align-items: center;
+        }}
+        .link-box {{
+            background: #161b22;
+            border: 1px solid #30363d;
+            border-radius: 6px;
+            padding: 15px 30px;
+            min-width: 300px;
+            text-align: center;
+            transition: all 0.3s ease;
+        }}
+        .link-box:hover {{
+            background: #1c2128;
+            border-color: #58a6ff;
+            transform: translateY(-2px);
+            box-shadow: 0 4px 12px rgba(88, 166, 255, 0.2);
+        }}
+        a {{
+            color: #58a6ff;
+            text-decoration: none;
+            font-size: 20px;
+            font-weight: 700;
+        }}
+        a:hover {{
+            color: #79c0ff;
+        }}
+        .canary-token {{
+            background: #1c1917;
+            border: 2px solid #f85149;
+            border-radius: 8px;
+            padding: 30px 50px;
+            margin: 40px auto;
+            max-width: 800px;
+            overflow-x: auto;
+        }}
+        .canary-token a {{
+            color: #f85149;
+            font-size: 18px;
+            white-space: nowrap;
+        }}
+    </style>
+</head>
+<body>
+    <div class="container">
+        <h1>Krawl me! &#128376;</h1>
+        <div class="counter">{Handler.counter}</div>
+        
+        <div class="links-container">
+"""
+
+        if Handler.counter <= 0 and self.config.canary_token_url:
+            html += f"""
+            <div class="link-box canary-token">
+                <a href="{self.config.canary_token_url}">{self.config.canary_token_url}</a>
+            </div>
+"""
+
+        if self.webpages is None:
+            for _ in range(num_pages):
+                address = ''.join([
+                    random.choice(self.config.char_space)
+                    for _ in range(random.randint(*self.config.links_length_range))
+                ])
+                html += f"""
+            <div class="link-box">
+                <a href="{address}">{address}</a>
+            </div>
+"""
+        else:
+            for _ in range(num_pages):
+                address = random.choice(self.webpages)
+                html += f"""
+            <div class="link-box">
+                <a href="{address}">{address}</a>
+            </div>
+"""
+
+        html += """
+        </div>
+    </div>
+</body>
+</html>"""
+        return html
+
+    def do_HEAD(self):
+        """Sends header information"""
+        self.send_response(200)
+        self.send_header("Content-type", "text/html")
+        self.end_headers()
+
+    def do_POST(self):
+        """Handle POST requests (mainly login attempts)"""
+        client_ip = self._get_client_ip()
+        user_agent = self._get_user_agent()
+        
+        self.tracker.record_access(client_ip, self.path, user_agent)
+        
+        print(f"[LOGIN ATTEMPT] {client_ip} - {self.path} - {user_agent[:50]}")
+        
+        content_length = int(self.headers.get('Content-Length', 0))
+        if content_length > 0:
+            post_data = self.rfile.read(content_length).decode('utf-8')
+            print(f"[POST DATA] {post_data[:200]}")
+        
+        time.sleep(1)
+        
+        self.send_response(200)
+        self.send_header('Content-type', 'text/html')
+        self.end_headers()
+        self.wfile.write(html_templates.login_error().encode())
+
+    def serve_special_path(self, path: str) -> bool:
+        """Serve special paths like robots.txt, API endpoints, etc."""
+        
+        if path == '/robots.txt':
+            self.send_response(200)
+            self.send_header('Content-type', 'text/plain')
+            self.end_headers()
+            self.wfile.write(html_templates.robots_txt().encode())
+            return True
+        
+        if path in ['/credentials.txt', '/passwords.txt', '/admin_notes.txt']:
+            self.send_response(200)
+            self.send_header('Content-type', 'text/plain')
+            self.end_headers()
+            if 'credentials' in path:
+                self.wfile.write(credentials_txt().encode())
+            else:
+                self.wfile.write(passwords_txt().encode())
+            return True
+        
+        if path in ['/users.json', '/api_keys.json', '/config.json']:
+            self.send_response(200)
+            self.send_header('Content-type', 'application/json')
+            self.end_headers()
+            if 'users' in path:
+                self.wfile.write(users_json().encode())
+            elif 'api_keys' in path:
+                self.wfile.write(api_keys_json().encode())
+            else:
+                self.wfile.write(api_response('/api/config').encode())
+            return True
+        
+        if path in ['/admin', '/admin/', '/admin/login', '/login', '/wp-login.php']:
+            self.send_response(200)
+            self.send_header('Content-type', 'text/html')
+            self.end_headers()
+            self.wfile.write(html_templates.login_form().encode())
+            return True
+        
+        if path == '/wp-admin' or path == '/wp-admin/':
+            self.send_response(200)
+            self.send_header('Content-type', 'text/html')
+            self.end_headers()
+            self.wfile.write(html_templates.login_form().encode())
+            return True
+        
+        if path in ['/wp-content/', '/wp-includes/'] or 'wordpress' in path.lower():
+            self.send_response(200)
+            self.send_header('Content-type', 'text/html')
+            self.end_headers()
+            self.wfile.write(html_templates.wordpress().encode())
+            return True
+        
+        if 'phpmyadmin' in path.lower() or path in ['/pma/', '/phpMyAdmin/']:
+            self.send_response(200)
+            self.send_header('Content-type', 'text/html')
+            self.end_headers()
+            self.wfile.write(html_templates.phpmyadmin().encode())
+            return True
+        
+        if path.startswith('/api/') or path.startswith('/api') or path in ['/.env']:
+            self.send_response(200)
+            self.send_header('Content-type', 'application/json')
+            self.end_headers()
+            self.wfile.write(api_response(path).encode())
+            return True
+        
+        if path in ['/backup/', '/uploads/', '/private/', '/admin/', '/config/', '/database/']:
+            self.send_response(200)
+            self.send_header('Content-type', 'text/html')
+            self.end_headers()
+            self.wfile.write(directory_listing(path).encode())
+            return True
+        
+        return False
+
+    def do_GET(self):
+        """Responds to webpage requests"""
+        client_ip = self._get_client_ip()
+        user_agent = self._get_user_agent()
+        
+        if self.config.dashboard_secret_path and self.path == self.config.dashboard_secret_path:
+            self.send_response(200)
+            self.send_header('Content-type', 'text/html')
+            self.end_headers()
+            try:
+                stats = self.tracker.get_stats()
+                self.wfile.write(generate_dashboard(stats).encode())
+            except Exception as e:
+                print(f"Error generating dashboard: {e}")
+            return
+
+        self.tracker.record_access(client_ip, self.path, user_agent)
+
+        if self.tracker.is_suspicious_user_agent(user_agent):
+            print(f"[SUSPICIOUS] {client_ip} - {user_agent[:50]} - {self.path}")
+
+        if self._should_return_error():
+            error_code = self._get_random_error_code()
+            print(f"[ERROR] Returning {error_code} to {client_ip} - {self.path}")
+            self.send_response(error_code)
+            self.end_headers()
+            return
+
+        if self.serve_special_path(self.path):
+            return
+
+        time.sleep(self.config.delay / 1000.0)
+        self.send_response(200)
+        self.send_header('Content-type', 'text/html')
+        self.end_headers()
+
+        try:
+            self.wfile.write(self.generate_page(self.path).encode())
+            
+            Handler.counter -= 1
+            
+            if Handler.counter < 0:
+                Handler.counter = self.config.canary_token_tries
+        except Exception as e:
+            print(f"Error generating page: {e}")
+
+    def log_message(self, format, *args):
+        """Override to customize logging"""
+        client_ip = self._get_client_ip()
+        print(f"[{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}] {client_ip} - {format % args}")

src/server.py

@@ -0,0 +1,83 @@
+#!/usr/bin/env python3
+
+"""
+Main server module for the deception honeypot.
+Run this file to start the server.
+"""
+
+import sys
+from http.server import HTTPServer
+
+from config import Config
+from tracker import AccessTracker
+from handler import Handler
+
+
+def print_usage():
+    """Print usage information"""
+    print(f'Usage: {sys.argv[0]} [FILE]\n')
+    print('FILE is file containing a list of webpage names to serve, one per line.')
+    print('If no file is provided, random links will be generated.\n')
+    print('Environment Variables:')
+    print('  PORT                  - Server port (default: 5000)')
+    print('  DELAY                 - Response delay in ms (default: 100)')
+    print('  LINKS_MIN_LENGTH      - Min link length (default: 5)')
+    print('  LINKS_MAX_LENGTH      - Max link length (default: 15)')
+    print('  LINKS_MIN_PER_PAGE    - Min links per page (default: 10)')
+    print('  LINKS_MAX_PER_PAGE    - Max links per page (default: 15)')
+    print('  MAX_COUNTER           - Max counter value (default: 10)')
+    print('  CANARY_TOKEN_URL      - Canary token URL to display')
+    print('  CANARY_TOKEN_TRIES    - Number of tries before showing token (default: 10)')
+    print('  DASHBOARD_SECRET_PATH - Secret path for dashboard (auto-generated if not set)')
+    print('  PROBABILITY_ERROR_CODES - Probability (0-100) to return HTTP error codes (default: 0)')
+    print('  CHAR_SPACE            - Characters for random links')
+
+
+def main():
+    """Main entry point for the deception server"""
+    if '-h' in sys.argv or '--help' in sys.argv:
+        print_usage()
+        exit(0)
+
+    config = Config.from_env()
+    
+    tracker = AccessTracker()
+    
+    Handler.config = config
+    Handler.tracker = tracker
+    Handler.counter = config.canary_token_tries
+
+    if len(sys.argv) == 2:
+        try:
+            with open(sys.argv[1], 'r') as f:
+                Handler.webpages = f.readlines()
+
+            if not Handler.webpages:
+                print('The file provided was empty. Using randomly generated links.')
+                Handler.webpages = None
+        except IOError:
+            print('Can\'t read input file. Using randomly generated links.')
+
+    try:
+        print(f'Starting deception server on port {config.port}...')
+        print(f'Dashboard available at: {config.dashboard_secret_path}')
+        if config.canary_token_url:
+            print(f'Canary token will appear after {config.canary_token_tries} tries')
+        else:
+            print('No canary token configured (set CANARY_TOKEN_URL to enable)')
+        
+        server = HTTPServer(('0.0.0.0', config.port), Handler)
+        print('Server started. Use <Ctrl-C> to stop.')
+        server.serve_forever()
+    except KeyboardInterrupt:
+        print('\nStopping server...')
+        server.socket.close()
+        print('Server stopped')
+    except Exception as e:
+        print(f'Error starting HTTP server on port {config.port}: {e}')
+        print(f'Make sure you are root, if needed, and that port {config.port} is open.')
+        exit(1)
+
+
+if __name__ == '__main__':
+    main()

src/templates/dashboard_template.py

@@ -0,0 +1,241 @@
+#!/usr/bin/env python3
+
+"""
+Dashboard template for viewing honeypot statistics.
+Customize this template to change the dashboard appearance.
+"""
+
+
+def generate_dashboard(stats: dict) -> str:
+    """Generate dashboard HTML with access statistics"""
+    
+    # Generate IP rows
+    top_ips_rows = '\n'.join([
+        f'<tr><td class="rank">{i+1}</td><td>{ip}</td><td>{count}</td></tr>'
+        for i, (ip, count) in enumerate(stats['top_ips'])
+    ]) or '<tr><td colspan="3" style="text-align:center;">No data</td></tr>'
+
+    # Generate paths rows
+    top_paths_rows = '\n'.join([
+        f'<tr><td class="rank">{i+1}</td><td>{path}</td><td>{count}</td></tr>'
+        for i, (path, count) in enumerate(stats['top_paths'])
+    ]) or '<tr><td colspan="3" style="text-align:center;">No data</td></tr>'
+
+    # Generate User-Agent rows
+    top_ua_rows = '\n'.join([
+        f'<tr><td class="rank">{i+1}</td><td style="word-break: break-all;">{ua[:80]}</td><td>{count}</td></tr>'
+        for i, (ua, count) in enumerate(stats['top_user_agents'])
+    ]) or '<tr><td colspan="3" style="text-align:center;">No data</td></tr>'
+
+    # Generate suspicious accesses rows
+    suspicious_rows = '\n'.join([
+        f'<tr><td>{log["ip"]}</td><td>{log["path"]}</td><td style="word-break: break-all;">{log["user_agent"][:60]}</td><td>{log["timestamp"].split("T")[1][:8]}</td></tr>'
+        for log in stats['recent_suspicious'][-10:]
+    ]) or '<tr><td colspan="4" style="text-align:center;">No suspicious activity detected</td></tr>'
+
+    # Generate honeypot triggered IPs rows
+    honeypot_rows = '\n'.join([
+        f'<tr><td>{ip}</td><td style="word-break: break-all;">{", ".join(paths)}</td><td>{len(paths)}</td></tr>'
+        for ip, paths in stats.get('honeypot_triggered_ips', [])
+    ]) or '<tr><td colspan="3" style="text-align:center;">No honeypot triggers yet</td></tr>'
+
+    return f"""<!DOCTYPE html>
+<html>
+<head>
+    <meta charset="UTF-8">
+    <title>Krawl Dashboard</title>
+    <style>
+        body {{
+            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+            background-color: #0d1117;
+            color: #c9d1d9;
+            margin: 0;
+            padding: 20px;
+        }}
+        .container {{
+            max-width: 1400px;
+            margin: 0 auto;
+        }}
+        h1 {{
+            color: #58a6ff;
+            text-align: center;
+            margin-bottom: 40px;
+        }}
+        .stats-grid {{
+            display: grid;
+            grid-template-columns: repeat(auto-fit, minmax(200px, 1fr));
+            gap: 20px;
+            margin-bottom: 40px;
+        }}
+        .stat-card {{
+            background: #161b22;
+            border: 1px solid #30363d;
+            border-radius: 6px;
+            padding: 20px;
+            text-align: center;
+        }}
+        .stat-card.alert {{
+            border-color: #f85149;
+        }}
+        .stat-value {{
+            font-size: 36px;
+            font-weight: bold;
+            color: #58a6ff;
+        }}
+        .stat-value.alert {{
+            color: #f85149;
+        }}
+        .stat-label {{
+            font-size: 14px;
+            color: #8b949e;
+            margin-top: 5px;
+        }}
+        .table-container {{
+            background: #161b22;
+            border: 1px solid #30363d;
+            border-radius: 6px;
+            padding: 20px;
+            margin-bottom: 20px;
+        }}
+        h2 {{
+            color: #58a6ff;
+            margin-top: 0;
+        }}
+        table {{
+            width: 100%;
+            border-collapse: collapse;
+        }}
+        th, td {{
+            padding: 12px;
+            text-align: left;
+            border-bottom: 1px solid #30363d;
+        }}
+        th {{
+            background: #0d1117;
+            color: #58a6ff;
+            font-weight: 600;
+        }}
+        tr:hover {{
+            background: #1c2128;
+        }}
+        .rank {{
+            color: #8b949e;
+            font-weight: bold;
+        }}
+        .alert-section {{
+            background: #1c1917;
+            border-left: 4px solid #f85149;
+        }}
+    </style>
+</head>
+<body>
+    <div class="container">
+        <h1>&#128375;&#65039; Krawl Dashboard</h1>
+        
+        <div class="stats-grid">
+            <div class="stat-card">
+                <div class="stat-value">{stats['total_accesses']}</div>
+                <div class="stat-label">Total Accesses</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{stats['unique_ips']}</div>
+                <div class="stat-label">Unique IPs</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{stats['unique_paths']}</div>
+                <div class="stat-label">Unique Paths</div>
+            </div>
+            <div class="stat-card alert">
+                <div class="stat-value alert">{stats['suspicious_accesses']}</div>
+                <div class="stat-label">Suspicious Accesses</div>
+            </div>
+            <div class="stat-card alert">
+                <div class="stat-value alert">{stats.get('honeypot_ips', 0)}</div>
+                <div class="stat-label">Honeypot Caught</div>
+            </div>
+        </div>
+
+        <div class="table-container alert-section">
+            <h2>🍯 Honeypot Triggers</h2>
+            <table>
+                <thead>
+                    <tr>
+                        <th>IP Address</th>
+                        <th>Accessed Paths</th>
+                        <th>Count</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {honeypot_rows}
+                </tbody>
+            </table>
+        </div>
+
+        <div class="table-container alert-section">
+            <h2>&#9888;&#65039; Recent Suspicious Activity</h2>
+            <table>
+                <thead>
+                    <tr>
+                        <th>IP Address</th>
+                        <th>Path</th>
+                        <th>User-Agent</th>
+                        <th>Time</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {suspicious_rows}
+                </tbody>
+            </table>
+        </div>
+
+        <div class="table-container">
+            <h2>Top IP Addresses</h2>
+            <table>
+                <thead>
+                    <tr>
+                        <th>#</th>
+                        <th>IP Address</th>
+                        <th>Access Count</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {top_ips_rows}
+                </tbody>
+            </table>
+        </div>
+
+        <div class="table-container">
+            <h2>Top Paths</h2>
+            <table>
+                <thead>
+                    <tr>
+                        <th>#</th>
+                        <th>Path</th>
+                        <th>Access Count</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {top_paths_rows}
+                </tbody>
+            </table>
+        </div>
+
+        <div class="table-container">
+            <h2>Top User-Agents</h2>
+            <table>
+                <thead>
+                    <tr>
+                        <th>#</th>
+                        <th>User-Agent</th>
+                        <th>Count</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {top_ua_rows}
+                </tbody>
+            </table>
+        </div>
+    </div>
+</body>
+</html>
+"""

src/templates/html_templates.py

@@ -0,0 +1,229 @@
+#!/usr/bin/env python3
+
+"""
+HTML templates for the deception server.
+Edit these templates to customize the appearance of fake pages.
+"""
+
+
+def login_form() -> str:
+    """Generate fake login page"""
+    return """<!DOCTYPE html>
+<html>
+<head>
+    <meta charset="UTF-8">
+    <title>Admin Login</title>
+    <style>
+        body { font-family: Arial, sans-serif; background: #f0f0f0; display: flex; justify-content: center; align-items: center; height: 100vh; margin: 0; }
+        .login-box { background: white; padding: 40px; border-radius: 8px; box-shadow: 0 2px 10px rgba(0,0,0,0.1); width: 300px; }
+        h2 { margin-top: 0; color: #333; }
+        input { width: 100%; padding: 10px; margin: 10px 0; border: 1px solid #ddd; border-radius: 4px; box-sizing: border-box; }
+        button { width: 100%; padding: 10px; background: #007bff; color: white; border: none; border-radius: 4px; cursor: pointer; }
+        button:hover { background: #0056b3; }
+    </style>
+</head>
+<body>
+    <div class="login-box">
+        <h2>Admin Login</h2>
+        <form action="/admin/login" method="post">
+            <input type="text" name="username" placeholder="Username" required>
+            <input type="password" name="password" placeholder="Password" required>
+            <button type="submit">Login</button>
+        </form>
+    </div>
+</body>
+</html>"""
+
+
+def login_error() -> str:
+    """Generate fake login error page"""
+    return """<!DOCTYPE html>
+<html>
+<head>
+    <meta charset="UTF-8">
+    <title>Login Failed</title>
+    <style>
+        body { font-family: Arial, sans-serif; background: #f0f0f0; display: flex; justify-content: center; align-items: center; height: 100vh; margin: 0; }
+        .login-box { background: white; padding: 40px; border-radius: 8px; box-shadow: 0 2px 10px rgba(0,0,0,0.1); width: 300px; }
+        h2 { margin-top: 0; color: #333; }
+        .error { color: #d63301; background: #ffebe8; border: 1px solid #d63301; padding: 12px; margin-bottom: 20px; border-radius: 4px; }
+        input { width: 100%; padding: 10px; margin: 10px 0; border: 1px solid #ddd; border-radius: 4px; box-sizing: border-box; }
+        button { width: 100%; padding: 10px; background: #007bff; color: white; border: none; border-radius: 4px; cursor: pointer; }
+        button:hover { background: #0056b3; }
+        a { color: #007bff; font-size: 14px; }
+    </style>
+</head>
+<body>
+    <div class="login-box">
+        <h2>Admin Login</h2>
+        <div class="error"><strong>ERROR:</strong> Invalid username or password.</div>
+        <form action="/admin/login" method="post">
+            <input type="text" name="username" placeholder="Username" required>
+            <input type="password" name="password" placeholder="Password" required>
+            <button type="submit">Login</button>
+        </form>
+        <p style="margin-top: 20px; text-align: center;"><a href="/forgot-password">Forgot your password?</a></p>
+    </div>
+</body>
+</html>"""
+
+
+def wordpress() -> str:
+    """Generate fake WordPress page"""
+    return """<!DOCTYPE html>
+<html lang="en-US">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <title>My Blog &#8211; Just another WordPress site</title>
+    <link rel='dns-prefetch' href='//s.w.org' />
+    <link rel='stylesheet' id='wp-block-library-css' href='/wp-includes/css/dist/block-library/style.min.css' type='text/css' media='all' />
+    <link rel='stylesheet' id='twentytwentythree-style-css' href='/wp-content/themes/twentytwentythree/style.css' type='text/css' media='all' />
+    <link rel='https://api.w.org/' href='/wp-json/' />
+    <meta name="generator" content="WordPress 6.4.2" />
+    <style>
+        body { font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, Oxygen-Sans, Ubuntu, Cantarell, "Helvetica Neue", sans-serif; margin: 0; padding: 0; background: #fff; }
+        .site-header { background: #23282d; color: white; padding: 20px; border-bottom: 4px solid #0073aa; }
+        .site-header h1 { margin: 0; font-size: 28px; }
+        .site-header p { margin: 5px 0 0; color: #d0d0d0; }
+        .site-content { max-width: 1200px; margin: 40px auto; padding: 0 20px; }
+        .entry { background: #fff; margin-bottom: 40px; padding: 30px; border: 1px solid #ddd; border-radius: 4px; }
+        .entry-title { font-size: 32px; margin-top: 0; color: #23282d; }
+        .entry-meta { color: #666; font-size: 14px; margin-bottom: 20px; }
+        .entry-content { line-height: 1.8; color: #444; }
+        .site-footer { background: #f7f7f7; padding: 20px; text-align: center; color: #666; border-top: 1px solid #ddd; margin-top: 60px; }
+        a { color: #0073aa; text-decoration: none; }
+        a:hover { text-decoration: underline; }
+    </style>
+</head>
+<body class="home blog wp-embed-responsive">
+<div id="page" class="site">
+    <header id="masthead" class="site-header">
+        <div class="site-branding">
+            <h1 class="site-title">My Blog</h1>
+            <p class="site-description">Just another WordPress site</p>
+        </div>
+    </header>
+
+    <div id="content" class="site-content">
+        <article id="post-1" class="entry">
+            <header class="entry-header">
+                <h2 class="entry-title">Hello world!</h2>
+                <div class="entry-meta">
+                    <span class="posted-on">Posted on <time datetime="2024-12-01">December 1, 2024</time></span>
+                    <span class="byline"> by <span class="author">admin</span></span>
+                </div>
+            </header>
+            <div class="entry-content">
+                <p>Welcome to WordPress. This is your first post. Edit or delete it, then start writing!</p>
+                <p>Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.</p>
+            </div>
+        </article>
+
+        <article id="post-2" class="entry">
+            <header class="entry-header">
+                <h2 class="entry-title">About This Site</h2>
+                <div class="entry-meta">
+                    <span class="posted-on">Posted on <time datetime="2024-11-28">November 28, 2024</time></span>
+                    <span class="byline"> by <span class="author">admin</span></span>
+                </div>
+            </header>
+            <div class="entry-content">
+                <p>This is a sample page. You can use it to write about your site, yourself, or anything else you'd like.</p>
+            </div>
+        </article>
+    </div>
+
+    <footer id="colophon" class="site-footer">
+        <div class="site-info">
+            Proudly powered by <a href="https://wordpress.org/">WordPress</a>
+        </div>
+    </footer>
+</div>
+<script type='text/javascript' src='/wp-includes/js/wp-embed.min.js'></script>
+</body>
+</html>"""
+
+
+def phpmyadmin() -> str:
+    """Generate fake phpMyAdmin page"""
+    return """<!DOCTYPE html>
+<html>
+<head>
+    <title>phpMyAdmin</title>
+    <style>
+        body { font-family: 'Segoe UI', Tahoma, sans-serif; margin: 0; background: #f0f0f0; }
+        .header { background: #2979ff; color: white; padding: 10px 20px; }
+        .login { background: white; width: 400px; margin: 100px auto; padding: 30px; border-radius: 4px; box-shadow: 0 2px 4px rgba(0,0,0,0.1); }
+        input { width: 100%; padding: 8px; margin: 8px 0; border: 1px solid #ddd; }
+        button { padding: 10px 20px; background: #2979ff; color: white; border: none; cursor: pointer; }
+    </style>
+</head>
+<body>
+    <div class="header"><h1>phpMyAdmin</h1></div>
+    <div class="login">
+        <h2>MySQL Server Login</h2>
+        <form action="/phpMyAdmin/index.php" method="post">
+            <input type="text" name="pma_username" placeholder="Username">
+            <input type="password" name="pma_password" placeholder="Password">
+            <button type="submit">Go</button>
+        </form>
+    </div>
+</body>
+</html>"""
+
+
+def robots_txt() -> str:
+    """Generate juicy robots.txt"""
+    return """User-agent: *
+Disallow: /admin/
+Disallow: /api/
+Disallow: /backup/
+Disallow: /config/
+Disallow: /database/
+Disallow: /private/
+Disallow: /uploads/
+Disallow: /wp-admin/
+Disallow: /phpMyAdmin/
+Disallow: /admin/login.php
+Disallow: /api/v1/users
+Disallow: /api/v2/secrets
+Disallow: /.env
+Disallow: /credentials.txt
+Disallow: /passwords.txt
+Disallow: /.git/
+Disallow: /backup.sql
+Disallow: /db_backup.sql
+"""
+
+
+def directory_listing(path: str, dirs: list, files: list) -> str:
+    """Generate fake directory listing"""
+    html = f"""<!DOCTYPE html>
+<html>
+<head><title>Index of {path}</title>
+<style>
+    body {{ font-family: monospace; background: #fff; padding: 20px; }}
+    h1 {{ border-bottom: 1px solid #ccc; padding-bottom: 10px; }}
+    table {{ width: 100%; border-collapse: collapse; }}
+    th {{ text-align: left; padding: 10px; background: #f0f0f0; }}
+    td {{ padding: 8px; border-bottom: 1px solid #eee; }}
+    a {{ color: #0066cc; text-decoration: none; }}
+    a:hover {{ text-decoration: underline; }}
+</style>
+</head>
+<body>
+<h1>Index of {path}</h1>
+<table>
+<tr><th>Name</th><th>Last Modified</th><th>Size</th></tr>
+<tr><td><a href="../">Parent Directory</a></td><td>-</td><td>-</td></tr>
+"""
+    
+    for d in dirs:
+        html += f'<tr><td><a href="{d}">{d}</a></td><td>2024-12-01 10:30</td><td>-</td></tr>\n'
+    
+    for f, size in files:
+        html += f'<tr><td><a href="{f}">{f}</a></td><td>2024-12-01 14:22</td><td>{size}</td></tr>\n'
+    
+    html += '</table></body></html>'
+    return html

src/tracker.py

@@ -0,0 +1,114 @@
+#!/usr/bin/env python3
+
+from typing import Dict, List, Tuple
+from collections import defaultdict
+from datetime import datetime
+
+
+class AccessTracker:
+    """Track IP addresses and paths accessed"""
+    def __init__(self):
+        self.ip_counts: Dict[str, int] = defaultdict(int)
+        self.path_counts: Dict[str, int] = defaultdict(int)
+        self.user_agent_counts: Dict[str, int] = defaultdict(int)
+        self.access_log: List[Dict] = []
+        self.suspicious_patterns = [
+            'bot', 'crawler', 'spider', 'scraper', 'curl', 'wget', 'python-requests',
+            'scanner', 'nikto', 'sqlmap', 'nmap', 'masscan', 'nessus', 'acunetix',
+            'burp', 'zap', 'w3af', 'metasploit', 'nuclei', 'gobuster', 'dirbuster'
+        ]
+        # Track IPs that accessed honeypot paths from robots.txt
+        self.honeypot_triggered: Dict[str, List[str]] = defaultdict(list)
+
+    def record_access(self, ip: str, path: str, user_agent: str = ''):
+        """Record an access attempt"""
+        self.ip_counts[ip] += 1
+        self.path_counts[path] += 1
+        if user_agent:
+            self.user_agent_counts[user_agent] += 1
+        
+        is_suspicious = self.is_suspicious_user_agent(user_agent) or self.is_honeypot_path(path)
+        
+        # Track if this IP accessed a honeypot path
+        if self.is_honeypot_path(path):
+            self.honeypot_triggered[ip].append(path)
+        
+        self.access_log.append({
+            'ip': ip,
+            'path': path,
+            'user_agent': user_agent,
+            'suspicious': is_suspicious,
+            'honeypot_triggered': self.is_honeypot_path(path),
+            'timestamp': datetime.now().isoformat()
+        })
+
+    def is_honeypot_path(self, path: str) -> bool:
+        """Check if path is one of the honeypot traps from robots.txt"""
+        honeypot_paths = [
+            '/admin',
+            '/admin/',
+            '/backup',
+            '/backup/',
+            '/config',
+            '/config/',
+            '/private',
+            '/private/',
+            '/database',
+            '/database/',
+            '/credentials.txt',
+            '/passwords.txt',
+            '/admin_notes.txt',
+            '/api_keys.json',
+            '/.env',
+            '/wp-admin',
+            '/wp-admin/',
+            '/phpmyadmin',
+            '/phpMyAdmin/'
+        ]
+        return path in honeypot_paths or any(hp in path.lower() for hp in ['/backup', '/admin', '/config', '/private', '/database', 'phpmyadmin'])
+
+    def is_suspicious_user_agent(self, user_agent: str) -> bool:
+        """Check if user agent matches suspicious patterns"""
+        if not user_agent:
+            return True
+        ua_lower = user_agent.lower()
+        return any(pattern in ua_lower for pattern in self.suspicious_patterns)
+
+    def get_top_ips(self, limit: int = 10) -> List[Tuple[str, int]]:
+        """Get top N IP addresses by access count"""
+        return sorted(self.ip_counts.items(), key=lambda x: x[1], reverse=True)[:limit]
+
+    def get_top_paths(self, limit: int = 10) -> List[Tuple[str, int]]:
+        """Get top N paths by access count"""
+        return sorted(self.path_counts.items(), key=lambda x: x[1], reverse=True)[:limit]
+
+    def get_top_user_agents(self, limit: int = 10) -> List[Tuple[str, int]]:
+        """Get top N user agents by access count"""
+        return sorted(self.user_agent_counts.items(), key=lambda x: x[1], reverse=True)[:limit]
+
+    def get_suspicious_accesses(self, limit: int = 20) -> List[Dict]:
+        """Get recent suspicious accesses"""
+        suspicious = [log for log in self.access_log if log.get('suspicious', False)]
+        return suspicious[-limit:]
+
+    def get_honeypot_triggered_ips(self) -> List[Tuple[str, List[str]]]:
+        """Get IPs that accessed honeypot paths"""
+        return [(ip, paths) for ip, paths in self.honeypot_triggered.items()]
+
+    def get_stats(self) -> Dict:
+        """Get statistics summary"""
+        suspicious_count = sum(1 for log in self.access_log if log.get('suspicious', False))
+        honeypot_count = sum(1 for log in self.access_log if log.get('honeypot_triggered', False))
+        return {
+            'total_accesses': len(self.access_log),
+            'unique_ips': len(self.ip_counts),
+            'unique_paths': len(self.path_counts),
+            'suspicious_accesses': suspicious_count,
+            'honeypot_triggered': honeypot_count,
+            'honeypot_ips': len(self.honeypot_triggered),
+            'top_ips': self.get_top_ips(10),
+            'top_paths': self.get_top_paths(10),
+            'top_user_agents': self.get_top_user_agents(10),
+            'recent_suspicious': self.get_suspicious_accesses(20),
+            'honeypot_triggered_ips': self.get_honeypot_triggered_ips()
+        }

src/wordlists.py

@@ -0,0 +1,123 @@
+#!/usr/bin/env python3
+
+"""
+Wordlists loader - reads all wordlists from wordlists.json
+This allows easy customization without touching Python code.
+"""
+
+import json
+import os
+from pathlib import Path
+
+
+class Wordlists:
+    """Loads and provides access to wordlists from wordlists.json"""
+    
+    def __init__(self):
+        self._data = self._load_config()
+    
+    def _load_config(self):
+        """Load wordlists from JSON file"""
+        config_path = Path(__file__).parent.parent / 'wordlists.json'
+        
+        try:
+            with open(config_path, 'r') as f:
+                return json.load(f)
+        except FileNotFoundError:
+            print(f"⚠️  Warning: {config_path} not found, using default values")
+            return self._get_defaults()
+        except json.JSONDecodeError as e:
+            print(f"⚠️  Warning: Invalid JSON in {config_path}: {e}")
+            return self._get_defaults()
+    
+    def _get_defaults(self):
+        """Fallback default wordlists if JSON file is missing or invalid"""
+        return {
+            "usernames": {
+                "prefixes": ["admin", "user", "root"],
+                "suffixes": ["", "_prod", "_dev"]
+            },
+            "passwords": {
+                "prefixes": ["P@ssw0rd", "Admin"],
+                "simple": ["test", "demo", "password"]
+            },
+            "emails": {
+                "domains": ["example.com", "test.com"]
+            },
+            "api_keys": {
+                "prefixes": ["sk_live_", "api_", ""]
+            },
+            "databases": {
+                "names": ["production", "main_db"],
+                "hosts": ["localhost", "db.internal"]
+            },
+            "applications": {
+                "names": ["WebApp", "Dashboard"]
+            },
+            "users": {
+                "roles": ["Administrator", "User"]
+            }
+        }
+    
+    @property
+    def username_prefixes(self):
+        return self._data.get("usernames", {}).get("prefixes", [])
+    
+    @property
+    def username_suffixes(self):
+        return self._data.get("usernames", {}).get("suffixes", [])
+    
+    @property
+    def password_prefixes(self):
+        return self._data.get("passwords", {}).get("prefixes", [])
+    
+    @property
+    def simple_passwords(self):
+        return self._data.get("passwords", {}).get("simple", [])
+    
+    @property
+    def email_domains(self):
+        return self._data.get("emails", {}).get("domains", [])
+    
+    @property
+    def api_key_prefixes(self):
+        return self._data.get("api_keys", {}).get("prefixes", [])
+    
+    @property
+    def database_names(self):
+        return self._data.get("databases", {}).get("names", [])
+    
+    @property
+    def database_hosts(self):
+        return self._data.get("databases", {}).get("hosts", [])
+    
+    @property
+    def application_names(self):
+        return self._data.get("applications", {}).get("names", [])
+    
+    @property
+    def user_roles(self):
+        return self._data.get("users", {}).get("roles", [])
+    
+    @property
+    def directory_files(self):
+        return self._data.get("directory_listing", {}).get("files", [])
+    
+    @property
+    def directory_dirs(self):
+        return self._data.get("directory_listing", {}).get("directories", [])
+    
+    @property
+    def error_codes(self):
+        return self._data.get("error_codes", [])
+
+
+_wordlists_instance = None
+
+def get_wordlists():
+    """Get the singleton Wordlists instance"""
+    global _wordlists_instance
+    if _wordlists_instance is None:
+        _wordlists_instance = Wordlists()
+    return _wordlists_instance
+

wordlists.json

@@ -0,0 +1,197 @@
+{
+  "usernames": {
+    "prefixes": [
+      "admin",
+      "user",
+      "developer",
+      "root",
+      "system",
+      "db",
+      "api",
+      "service",
+      "deploy",
+      "test",
+      "prod",
+      "backup",
+      "monitor",
+      "jenkins",
+      "webapp"
+    ],
+    "suffixes": [
+      "",
+      "_prod",
+      "_dev",
+      "_test",
+      "123",
+      "2024",
+      "_backup",
+      "_admin",
+      "01",
+      "02",
+      "_user",
+      "_service",
+      "_api"
+    ]
+  },
+  "passwords": {
+    "prefixes": [
+      "P@ssw0rd",
+      "Passw0rd",
+      "Admin",
+      "Secret",
+      "Welcome",
+      "System",
+      "Database",
+      "Secure",
+      "Master",
+      "Root"
+    ],
+    "simple": [
+      "test",
+      "demo",
+      "temp",
+      "change",
+      "password",
+      "admin",
+      "letmein",
+      "welcome",
+      "default",
+      "sample"
+    ]
+  },
+  "emails": {
+    "domains": [
+      "example.com",
+      "company.com",
+      "localhost.com",
+      "test.com",
+      "domain.com",
+      "corporate.com",
+      "internal.net",
+      "enterprise.com",
+      "business.org"
+    ]
+  },
+  "api_keys": {
+    "prefixes": [
+      "sk_live_",
+      "sk_test_",
+      "api_",
+      "key_",
+      "token_",
+      "access_",
+      "secret_",
+      "prod_",
+      ""
+    ]
+  },
+  "databases": {
+    "names": [
+      "production",
+      "prod_db",
+      "main_db",
+      "app_database",
+      "users_db",
+      "customer_data",
+      "analytics",
+      "staging_db",
+      "dev_database",
+      "wordpress",
+      "ecommerce",
+      "crm_db",
+      "inventory"
+    ],
+    "hosts": [
+      "localhost",
+      "db.internal",
+      "mysql.local",
+      "postgres.internal",
+      "127.0.0.1",
+      "db-server-01",
+      "database.prod",
+      "sql.company.com"
+    ]
+  },
+  "applications": {
+    "names": [
+      "WebApp",
+      "API Gateway",
+      "Dashboard",
+      "Admin Panel",
+      "CMS",
+      "Portal",
+      "Manager",
+      "Console",
+      "Control Panel",
+      "Backend"
+    ]
+  },
+  "users": {
+    "roles": [
+      "Administrator",
+      "Developer",
+      "Manager",
+      "User",
+      "Guest",
+      "Moderator",
+      "Editor",
+      "Viewer",
+      "Analyst",
+      "Support"
+    ]
+  },
+  "directory_listing": {
+    "files": [
+      "test.exe",
+      "backup.sql",
+      "database.sql",
+      "db_backup.sql",
+      "dump.sql",
+      "config.php",
+      "credentials.txt",
+      "passwords.txt",
+      "users.csv",
+      ".env",
+      "id_rsa",
+      "id_rsa.pub",
+      "private_key.pem",
+      "api_keys.json",
+      "secrets.yaml",
+      "admin_notes.txt",
+      "settings.ini",
+      "database.yml",
+      "wp-config.php",
+      ".htaccess",
+      "server.key",
+      "cert.pem",
+      "shadow.bak",
+      "passwd.old"
+    ],
+    "directories": [
+      "uploads/",
+      "backups/",
+      "logs/",
+      "temp/",
+      "cache/",
+      "private/",
+      "config/",
+      "admin/",
+      "database/",
+      "backup/",
+      "old/",
+      "archive/",
+      ".git/",
+      "keys/",
+      "credentials/"
+    ]
+  },
+  "error_codes": [
+    400,
+    401,
+    403,
+    404,
+    500,
+    502,
+    503
+  ]
+}