DigiJ 67b7edc696 v1.5.1 — Fix chat system, add system tray icon, agent mode improvements

- Fix Hal chat: add Chat/Agent mode toggle so users can switch between
  direct LLM streaming (Chat) and tool-using Agent mode
- Fix Agent system: graceful degradation when model can't follow
  structured THOUGHT/ACTION/PARAMS format (falls back to direct answer
  after 2 parse failures instead of looping 20 times)
- Fix frozen build: remove llama_cpp from PyInstaller excludes list
  so LLM works in compiled exe
- Add system tray icon: autarch.ico (from icon.svg) used for exe icons,
  installer shortcuts, and runtime tray icon
- Update tray.py to load .ico file with fallback to programmatic generation
- Add inline critical CSS for FOUC prevention
- Bump version to 1.5.1

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-03-02 23:13:13 -08:00

164 KiB

Raw Blame History

AUTARCH Development Log

Project: darkHal Security Group - Project AUTARCH

Session 1 - 2026-01-14

Initial Request

User Prompt:

Hello claude. Today we are building the framework for the darkHal Security Group's "Project AUTARCH". Right now all need to do is setup the frameworks main menu's and the LLM intagration using llamacpp and llama-cpp-python (the python packages are installed). the core c files are https://github.com/ggml-org/llama.cpp. The framework will be written in python and besides running the LLM and the agent its only other function is to load python modules that do the work. In other words, this application is just the front end. Lets start with the main menu. The main menu should have a ascii banner that says "AUTARCH" with a tagline that says "By darkHal and Setec Security Labs." On first launch the application auto runs the setup module (we need to create this as well). The setup module will setup the autarch_settings.conf. This module controls the llamacpp settings

Work Completed

1. Project Structure Created

dh_framework/
├── autarch.py          # Main entry point
├── autarch_settings.conf  # Configuration file (auto-generated)
├── core/
│   ├── __init__.py
│   ├── banner.py       # ASCII banner display
│   ├── config.py       # Configuration handler
│   └── menu.py         # Main menu system
└── modules/
    ├── __init__.py
    └── setup.py        # First-time setup wizard

2. Components Implemented

core/banner.py

ASCII art banner for "AUTARCH"
Color support using ANSI codes
Tagline: "By darkHal and Setec Security Labs."
Helper functions for display and screen clearing

core/config.py

Configuration handler for autarch_settings.conf
Manages llama.cpp settings:
- model_path
- n_ctx (context size)
- n_threads
- n_gpu_layers
- temperature
- top_p
- top_k
- repeat_penalty
- max_tokens
- seed
First-run detection
Type-safe getters (get_int, get_float, get_bool)

core/menu.py

Main menu with options:
1. Load Module
2. List Modules
3. Settings
4. Run Setup
5. Exit
Dynamic module loading from modules/ directory
Module info display (name, description, author, version)
Status display showing current model

modules/setup.py

Interactive first-time setup wizard
Configures all llama.cpp parameters
Model path validation
Input validation with min/max bounds
Saves configuration to autarch_settings.conf

autarch.py

Main entry point
Auto-runs setup on first launch
Launches main menu

Files Created

File	Lines	Purpose
core/init.py	2	Package init
modules/init.py	2	Package init
core/banner.py	42	ASCII banner
core/config.py	137	Config management
core/menu.py	232	Menu system
modules/setup.py	234	Setup wizard
autarch.py	47	Entry point

Session 1 (Continued) - LLM Integration

User Request

yes lets do the llm integration

Work Completed

3. LLM Integration Added

core/llm.py - LLM Wrapper Class

Wraps llama-cpp-python for model interaction
Features:
- load_model() - Load GGUF models with config settings
- unload_model() - Free resources
- generate() - Text completion with streaming support
- chat() - Chat-style interaction with history
- _build_chat_prompt() - ChatML-style prompt formatting
- clear_history() / get_history() / set_history() - Conversation management
- get_model_info() - Model metadata
Global instance via get_llm()
Custom LLMError exception class

modules/chat.py - Interactive Chat Module

Full chat interface for the LLM
Commands:
- /help - Show available commands
- /clear - Clear conversation history
- /history - Show conversation history
- /info - Show model information
- /system <prompt> - Set system prompt
- /temp <value> - Set temperature
- /tokens <value> - Set max tokens
- /stream - Toggle streaming mode
- /exit - Exit chat
Streaming output support
Automatic model loading
Default system prompt for AUTARCH persona

Updated Project Structure

dh_framework/
├── autarch.py              # Main entry point
├── autarch_settings.conf   # Configuration file (auto-generated)
├── DEVLOG.md               # Development log
├── core/
│   ├── __init__.py
│   ├── banner.py           # ASCII banner display
│   ├── config.py           # Configuration handler
│   ├── llm.py              # LLM wrapper (NEW)
│   └── menu.py             # Main menu system
└── modules/
    ├── __init__.py
    ├── chat.py             # Chat interface (NEW)
    └── setup.py            # First-time setup wizard

Files Added This Session

File	Lines	Purpose
core/llm.py	~280	LLM wrapper for llama-cpp-python
modules/chat.py	~250	Interactive chat interface

Session 1 (Continued) - Agent System

User Request

yes lets do the agent system

Work Completed

4. Agent System Implemented

core/tools.py - Tool System

Tool and ToolParameter dataclasses for tool definitions
ToolRegistry class for managing available tools
Built-in tools:
- shell - Execute shell commands with timeout
- read_file - Read file contents
- write_file - Write content to files
- list_dir - List directory contents
- search_files - Glob pattern file search
- search_content - Grep-style content search
- task_complete - Signal task completion
- ask_user - Request user input
JSON schema generation for LLM tool descriptions
Global registry via get_tool_registry()

core/agent.py - Autonomous Agent

Agent class implementing the autonomous loop
AgentState enum: IDLE, THINKING, EXECUTING, WAITING_USER, COMPLETE, ERROR
AgentStep dataclass for recording step history
AgentResult dataclass for task results
Features:
- Structured response parsing (THOUGHT/ACTION/PARAMS format)
- Tool execution with error handling
- User interaction via ask_user tool
- Step limit (default 20) to prevent infinite loops
- Callbacks for step completion and state changes
- ChatML-style prompt building
System prompt instructs LLM on tool usage format

modules/agent.py - Agent Interface Module

Interactive task input loop
Commands:
- tools - Show available tools
- exit - Return to main menu
- help - Show help
Task execution with progress display
Result summary with success/failure status
Step count reporting

Updated Project Structure

dh_framework/
├── autarch.py              # Main entry point
├── autarch_settings.conf   # Configuration file (auto-generated)
├── DEVLOG.md               # Development log
├── core/
│   ├── __init__.py
│   ├── agent.py            # Autonomous agent (NEW)
│   ├── banner.py           # ASCII banner display
│   ├── config.py           # Configuration handler
│   ├── llm.py              # LLM wrapper
│   ├── menu.py             # Main menu system
│   └── tools.py            # Tool system (NEW)
└── modules/
    ├── __init__.py
    ├── agent.py            # Agent interface (NEW)
    ├── chat.py             # Chat interface
    └── setup.py            # First-time setup wizard

Files Added This Session

File	Lines	Purpose
core/tools.py	~290	Tool definitions and registry
core/agent.py	~320	Autonomous agent loop
modules/agent.py	~175	Agent user interface

Agent Response Format

The agent uses a structured format:

THOUGHT: [reasoning about what to do]
ACTION: [tool_name]
PARAMS: {"param1": "value1"}

Observations are fed back as:

OBSERVATION: [tool output]

User Request

lets first add a few other features. Lets add metasploit compatability so the framework can use metasploit modules, and then add the main menu: 1) Defense 2) Offense 3) Counter 4) Analyze 5) OSINT 6) Simulate 99) Settings 98) Exit

Work Completed

5. Metasploit Integration

core/msf.py - Metasploit RPC Client

MetasploitRPC class for MSF RPC communication
Uses msgpack for binary protocol
Features:
- connect() / disconnect() - Authentication
- list_modules() / search_modules() - Module discovery
- get_module_info() / get_module_options() - Module details
- execute_module() - Run exploits/auxiliary modules
- list_jobs() / stop_job() - Job management
- list_sessions() - Active session listing
- session_shell_read() / session_shell_write() - Session interaction
- run_console_command() - Direct console access
MSFManager class for settings management
Configuration stored in autarch_settings.conf [msf] section

Agent MSF Tools Added to core/tools.py:

msf_connect - Connect to MSF RPC
msf_search - Search for modules
msf_module_info - Get module details
msf_module_options - Get module options
msf_execute - Execute modules
msf_sessions - List active sessions
msf_session_command - Run commands in sessions
msf_console - Direct console commands

New Menu Structure:

  Main Menu
  ──────────────────────────────────────────────────
  [1]  Defense      - Defensive security tools
  [2]  Offense      - Penetration testing
  [3]  Counter      - Counter-intelligence
  [4]  Analyze      - Analysis & forensics
  [5]  OSINT        - Open source intelligence
  [6]  Simulate     - Attack simulation

  [99] Settings
  [98] Exit

Category System:

Modules now have CATEGORY attribute
Categories: defense, offense, counter, analyze, osint, simulate, core
Category submenus show only relevant modules
Color-coded by category

Settings Menu:

LLM Settings
Metasploit Settings (with connection test)
View All Settings
Run Setup Wizard

Status Line:

Shows current model name
Shows MSF connection status

Updated Project Structure

dh_framework/
├── autarch.py              # Main entry point
├── autarch_settings.conf   # Configuration file
├── DEVLOG.md               # Development log
├── core/
│   ├── __init__.py
│   ├── agent.py            # Autonomous agent
│   ├── banner.py           # ASCII banner display
│   ├── config.py           # Configuration handler
│   ├── llm.py              # LLM wrapper
│   ├── menu.py             # Main menu (UPDATED)
│   ├── msf.py              # Metasploit integration (NEW)
│   └── tools.py            # Tool system (UPDATED)
└── modules/
    ├── __init__.py
    ├── agent.py            # Agent interface (CATEGORY: core)
    ├── chat.py             # Chat interface (CATEGORY: core)
    └── setup.py            # First-time setup wizard

Files Added/Modified This Session

File	Lines	Purpose
core/msf.py	~380	Metasploit RPC integration
core/menu.py	~480	Updated with categories
core/tools.py	~500	Added MSF tools

Module Categories

Category	Description	Color
defense	Defensive security tools	Blue
offense	Penetration testing	Red
counter	Counter-intelligence	Magenta
analyze	Analysis & forensics	Cyan
osint	Open source intelligence	Green
simulate	Attack simulation	Yellow
core	Core framework modules	White

MSF Configuration (autarch_settings.conf)

[msf]
host = 127.0.0.1
port = 55553
username = msf
password =
ssl = true

Session 1 (Continued) - Simplified MSF Interface

User Request

lets simplify how metasploit modules can be used

Work Completed

7. Simplified Metasploit Module

modules/msf.py - User-Friendly MSF Interface (CATEGORY: offense)

Menu-driven interface:

  Metasploit Framework
  ──────────────────────────────────────────────────
  Status: Connected/Disconnected
  Module: current/module (if selected)

  [1] Search Modules
  [2] Use Module
  [3] Show Options
  [4] Set Option
  [5] Run Module

  [6] Sessions
  [7] Jobs

  [8] Console Command
  [9] Quick Scan (auxiliary/scanner)

  [0] Back to Main Menu

Features:

Search Modules - Search by keyword, grouped results by type
Use Module - Select by full path OR search and pick from list
Show Options - Display required/optional with current values
Set Option - Set individual options (RHOSTS, LHOST, etc.)
Run Module - Execute with confirmation
Sessions - List and interact with active sessions
Jobs - View and kill running jobs
Console Command - Direct MSF console access
Quick Scan - Pre-configured scanners:
- TCP Port Scanner
- SMB Version Scanner
- SSH Version Scanner
- HTTP Version Scanner
- FTP Version Scanner
- MS17-010 (EternalBlue) Check

Workflow Example:

1. Select "Search Modules" → search "eternalblue"
2. Select "Use Module" → pick from results
3. Select "Set Option" → set RHOSTS=192.168.1.100
4. Select "Show Options" → verify configuration
5. Select "Run Module" → execute exploit
6. Select "Sessions" → interact with shell

Updated Project Structure

dh_framework/
├── autarch.py
├── autarch_settings.conf
├── DEVLOG.md
├── core/
│   ├── agent.py
│   ├── banner.py
│   ├── config.py
│   ├── llm.py
│   ├── menu.py
│   ├── msf.py
│   └── tools.py
└── modules/
    ├── __init__.py
    ├── agent.py      # CATEGORY: core
    ├── chat.py       # CATEGORY: core
    ├── msf.py        # CATEGORY: offense (NEW)
    └── setup.py

Files Added This Session

File	Lines	Purpose
modules/msf.py	~420	Simplified MSF interface

Session 1 (Continued) - Category Modules

User Request

lets add modules for the other categories

Work Completed

8. Category Modules Added

modules/defender.py - Defense Module (CATEGORY: defense)

Full Security Audit
Firewall Check (iptables/ufw/firewalld)
SSH Hardening Check
Open Ports Scan
User Security Check (UID 0, empty passwords)
File Permissions Check
Service Audit
Fail2Ban/SELinux/AppArmor detection
Security score calculation

modules/counter.py - Counter-Intelligence Module (CATEGORY: counter)

Full Threat Scan
Suspicious Process Detection (known malware, hidden processes)
Network Analysis (suspicious ports, C2 connections)
Login Anomalies (brute force detection, failed logins)
File Integrity Monitoring (critical file changes, SUID binaries)
Scheduled Task Audit (cron job analysis)
Rootkit Detection (hidden files, kernel modules)

modules/analyze.py - Forensics Module (CATEGORY: analyze)

File Analysis (metadata, hashes, type detection)
String Extraction (URLs, IPs, emails, paths)
Hash Lookup (VirusTotal/Hybrid Analysis links)
Log Analysis (IP extraction, error patterns)
Hex Dump viewer
File Comparison

modules/recon.py - OSINT Module (CATEGORY: osint)

Domain Reconnaissance (DNS, WHOIS, subdomains via crt.sh)
IP Address Lookup (reverse DNS, geolocation, quick port scan)
Email Harvester
Subdomain Enumeration (certificate transparency + brute force)
Technology Detection (server, CMS, frontend frameworks)

modules/simulate.py - Attack Simulation Module (CATEGORY: simulate)

Password Audit (strength analysis, hash generation)
Port Scanner (TCP scan with service detection)
Banner Grabber
Payload Generator (XSS, SQLi, Command Injection, Path Traversal, SSTI)
Network Stress Test (controlled)

Updated Project Structure

dh_framework/
├── autarch.py
├── autarch_settings.conf
├── DEVLOG.md
├── core/
│   ├── __init__.py
│   ├── agent.py
│   ├── banner.py
│   ├── config.py
│   ├── llm.py
│   ├── menu.py
│   ├── msf.py
│   └── tools.py
└── modules/
    ├── __init__.py
    ├── agent.py       # CATEGORY: core
    ├── analyze.py     # CATEGORY: analyze (NEW)
    ├── chat.py        # CATEGORY: core
    ├── counter.py     # CATEGORY: counter (NEW)
    ├── defender.py    # CATEGORY: defense (NEW)
    ├── msf.py         # CATEGORY: offense
    ├── recon.py       # CATEGORY: osint (NEW)
    ├── setup.py
    └── simulate.py    # CATEGORY: simulate (NEW)

Files Added This Session

File	Lines	Purpose
modules/defender.py	~280	System hardening checks
modules/counter.py	~350	Threat detection
modules/analyze.py	~320	Forensics tools
modules/recon.py	~330	OSINT reconnaissance
modules/simulate.py	~310	Attack simulation

Module Summary by Category

Category	Module	Features
Defense	defender	Security audit, firewall, SSH, permissions
Offense	msf	Metasploit interface, exploits, sessions
Counter	counter	Threat detection, rootkit checks, anomalies
Analyze	analyze	File forensics, strings, hashes, logs
OSINT	recon	Email/username/phone/domain/IP lookup
Simulate	simulate	Port scan, password audit, payloads
Core	agent	Autonomous AI agent
Core	chat	Interactive LLM chat

Session 1 (Continued) - Expanded OSINT

User Request

lets add modules for OSINT as well as add social-analyzer intagration. For OSINT lets add the sub catagories for email, username, phone number

Work Completed

9. Expanded OSINT Module

modules/recon.py - Enhanced to v2.0

New subcategorized menu:

  OSINT & Reconnaissance
  social-analyzer: Available/Not installed
  ──────────────────────────────────────────────────

  Email
    [1] Email Lookup
    [2] Email Permutator

  Username
    [3] Username Lookup
    [4] Social Analyzer

  Phone
    [5] Phone Number Lookup

  Domain/IP
    [6] Domain Recon
    [7] IP Address Lookup
    [8] Subdomain Enum
    [9] Tech Detection

  [0] Back

Email OSINT Features:

Email format analysis
MX record verification
Breach check resources (HaveIBeenPwned, DeHashed, IntelX)
Disposable email detection
Gravatar lookup
Email permutation generator (first.last, flast, etc.)

Username OSINT Features:

Multi-platform check (17+ platforms)
Twitter/X, Instagram, Facebook, GitHub, Reddit, LinkedIn
TikTok, YouTube, Pinterest, Twitch, Steam, Spotify
Medium, Dev.to, HackerNews, Keybase, Telegram
HTTP status verification
social-analyzer integration for deep scanning

Phone OSINT Features:

Country code detection (12 countries)
Carrier lookup resources (NumVerify, Twilio)
Search resources (TrueCaller, Sync.me, SpyDialer, WhitePages)
Messaging app check (WhatsApp, Telegram, Signal)
Spam/scam database check

social-analyzer Integration:

Auto-detection of installation
Deep profile scanning across 300+ sites
JSON output parsing
Profile link extraction

Updated recon.py Stats

Feature	Lines	Description
Email OSINT	~90	Lookup, permutator
Username OSINT	~100	Platform check, social-analyzer
Phone OSINT	~60	Number analysis, resources
Domain/IP	~200	DNS, WHOIS, subdomains, tech
Total	~590	Expanded from ~330

Notes

Framework uses llama-cpp-python for LLM integration (package pre-installed)
Modules can define DESCRIPTION, AUTHOR, VERSION, CATEGORY attributes
All modules must have a run() function entry point
Chat uses ChatML format (<|im_start|> / <|im_end|>) for compatibility
Agent uses lower temperature (0.3) for more focused tool selection
MSF RPC requires msfrpcd running: msfrpcd -P password -S
social-analyzer: pip install social-analyzer

Session 1 (Continued) - Adult Site Username Scanner

User Request

i am not doing the preditor tool anymore. Just a username OSINT tool like social-analyzer

Work Completed

10. Adult Site Username Scanner

modules/adultscan.py - Adult Site OSINT (CATEGORY: osint)

Username scanner for adult-oriented platforms with parallel scanning:

  Adult Site Scanner
  Username OSINT for adult platforms
  Sites in database: 50+
  ──────────────────────────────────────────────────

  [1] Full Scan (all categories)
  [2] Fanfiction & Story Sites
  [3] Art & Creative Sites
  [4] Video & Streaming Sites
  [5] Forums & Communities
  [6] Dating & Social Sites
  [7] Gaming Related Sites
  [8] Custom Category Selection

  [9] List All Sites

  [0] Back

Site Categories:

Category	Sites	Examples
fanfiction	9	Archive of Our Own, FanFiction.net, FimFiction, Wattpad, Literotica, Hentai Foundry
art	10	DeviantArt, Fur Affinity, Newgrounds, Pixiv, Rule34, e621, Tumblr
video	8	Pornhub, XVideos, xHamster, Chaturbate, OnlyFans, Fansly, ManyVids
forums	6	Reddit, F-List, FetLife, Kink.com, BDSMLR, CollarSpace
dating	5	AdultFriendFinder, Ashley Madison, Grindr, Scruff, Recon
gaming	4	F95zone, LoversLab, ULMF, Nutaku

Features:

Parallel scanning with ThreadPoolExecutor (10 workers)
Two detection methods:
- status - HTTP status code check (200/301/302 = found, 404 = not found)
- content - Page content analysis for sites with custom error pages
Progress indicator during scan
Category selection (single, multiple, or all)
Results export to file
Color-coded output (green = found, yellow = possible/redirect)

Detection Flow:

def check_site(self, site_info, username):
    # 1. Format URL with username
    url = url_template.format(username)

    # 2. Use curl to get HTTP status
    cmd = f"curl -sI -o /dev/null -w '%{{http_code}}' -L --max-time {timeout} '{url}'"

    # 3. Interpret based on method
    if method == 'status':
        # 200 = found, 404 = not found
    else:
        # Content-based: 200 = possible match

Updated Project Structure

dh_framework/
├── autarch.py
├── autarch_settings.conf
├── DEVLOG.md
├── core/
│   ├── __init__.py
│   ├── agent.py
│   ├── banner.py
│   ├── config.py
│   ├── llm.py
│   ├── menu.py
│   ├── msf.py
│   └── tools.py
└── modules/
    ├── __init__.py
    ├── adultscan.py    # CATEGORY: osint (NEW)
    ├── agent.py        # CATEGORY: core
    ├── analyze.py      # CATEGORY: analyze
    ├── chat.py         # CATEGORY: core
    ├── counter.py      # CATEGORY: counter
    ├── defender.py     # CATEGORY: defense
    ├── msf.py          # CATEGORY: offense
    ├── recon.py        # CATEGORY: osint
    ├── setup.py
    └── simulate.py     # CATEGORY: simulate

Files Added This Session

File	Lines	Purpose
modules/adultscan.py	~365	Adult site username scanner

OSINT Module Summary

Module	Purpose	Features
recon.py	General OSINT	Email, username (mainstream), phone, domain, IP
adultscan.py	Adult sites	50+ adult/fanfiction/art platforms

Notes

Scanner respects site rate limits via 10-second timeout per request
Uses curl for HTTP requests (more reliable than Python requests for some sites)
Some sites use content-based detection due to custom 404 pages
Export format: plain text with site names and URLs

Session 1 (Continued) - Custom Site Management

User Request

for the Adult Site Scanner Module, lets add two options, Manually add website to the list with the default username pattern such as if i wanted to add fakeadult.com to the list i would add fakeadult.com/user/* where the star gets replaces by the username

Work Completed

11. Custom Site Management for Adult Scanner

modules/adultscan.py - Updated to v1.1

Added custom site management features:

New Menu Structure:

  Adult Site Scanner
  Sites in database: 50+ (X custom)
  ──────────────────────────────────────────────────

  Scan Categories:
  [1] Full Scan (all categories)
  [2] Fanfiction & Story Sites
  [3] Art & Creative Sites
  [4] Video & Streaming Sites
  [5] Forums & Communities
  [6] Dating & Social Sites
  [7] Gaming Related Sites
  [8] Custom Sites Only
  [9] Custom Category Selection

  Site Management:
  [A] Add Custom Site
  [M] Manage Custom Sites
  [L] List All Sites

  [0] Back

Add Custom Site ([A]):

Prompts for site name
URL pattern using * as username placeholder
- Example: https://example.com/user/*
- Example: example.com/profile?name=*
Auto-adds https:// if no protocol specified
Detection method selection:
- Status code (default) - checks HTTP response
- Content - for sites with custom 404 pages
Saves to custom_adultsites.json

Manage Custom Sites ([M]):

Lists all custom sites with URL patterns and methods
Add new sites
Remove existing sites by number

Custom Sites Only ([8]):

Scan only user-added custom sites

Storage:

Custom sites stored in custom_adultsites.json in framework root
JSON format: {"sites": [["name", "url_template", "method"], ...]}
Persists between sessions

Example Usage:

Site name: FakeAdult
URL pattern (use * for username): fakeadult.com/user/*
Detection Method: [1] Status code

[+] Added 'FakeAdult' to custom sites
    URL: https://fakeadult.com/user/<username>

Files Modified

File	Changes
modules/adultscan.py	Added custom site management (~150 new lines)

New Methods Added

Method	Purpose
`load_custom_sites()`	Load from JSON file
`save_custom_sites()`	Save to JSON file
`add_custom_site()`	Interactive add wizard
`manage_custom_sites()`	View/manage menu
`remove_custom_site()`	Remove by index

Storage Format (custom_adultsites.json)

{
  "sites": [
    ["Site Name", "https://example.com/user/{}", "status"],
    ["Another Site", "https://other.com/profile/{}", "content"]
  ]
}

Note: * in user input is converted to {} for internal template formatting.

Session 1 (Continued) - Auto-Detect Site Patterns

User Request

lets add auto detection add, all the user has to is add fakeadult.com and the application just searches for a username using the most common patterns like fakeadult.com/u/* fakeadult.com/user/* etc

Work Completed

12. Auto-Detect Site Pattern Feature

modules/adultscan.py - Updated to v1.2

Added auto-detection that probes common URL patterns:

New Menu Option:

  Site Management:
  [A] Add Custom Site (manual)
  [D] Auto-Detect Site Pattern    <- NEW
  [M] Manage Custom Sites
  [L] List All Sites

Common Patterns Tested:

COMMON_PATTERNS = [
    '/user/{}',
    '/users/{}',
    '/u/{}',
    '/profile/{}',
    '/profiles/{}',
    '/member/{}',
    '/members/{}',
    '/@{}',
    '/{}',
    '/people/{}',
    '/account/{}',
    '/id/{}',
    '/{}/profile',
    '/user/{}/profile',
    '/channel/{}',
    '/c/{}',
    '/p/{}',
]

Workflow:

User enters just the domain (e.g., example.com)
User provides a known-existing username for testing
System probes all 17 common patterns
Shows which patterns return 200/301/302 responses
User selects the working pattern to add
Site is saved to custom sites

Example Usage:

Domain: fakeadult.com
Test username: knownuser

Testing 17 common URL patterns...

Found 2 working pattern(s):

  [1] /user/{}            (OK)
      https://fakeadult.com/user/knownuser
  [2] /u/{}               (redirect 302)
      https://fakeadult.com/u/knownuser

  [0] Cancel

Select pattern to add: 1
Site name [Fakeadult]: FakeAdult
[+] Added 'FakeAdult' to custom sites
    Pattern: https://fakeadult.com/user/*

Detection Logic:

Uses 5-second timeout per pattern
HTTP 200 = confirmed working
HTTP 301/302 = working (redirect)
HTTP 404/other = not working
Auto-selects detection method based on response

Files Modified

File	Changes
modules/adultscan.py	Added auto-detect (~90 new lines)

New Components

Component	Purpose
`COMMON_PATTERNS`	List of 17 common URL patterns
`auto_detect_site()`	Interactive auto-detection wizard

Notes

Auto-detect requires a known valid username to test against
Some sites may have rate limiting that affects detection
Falls back to manual add if no patterns work

Session 1 (Continued) - Bulk Import Feature

User Request

lets also add a bulk list function where the user can add sites to custom.inf and then it scans for new sites using autodetect

Work Completed

13. Bulk Import from File

modules/adultscan.py - Updated to v1.3

Added bulk import that reads domains from a file and auto-detects patterns:

New Menu Option:

  Site Management:
  [A] Add Custom Site (manual)
  [D] Auto-Detect Site Pattern
  [B] Bulk Import from File        <- NEW
  [M] Manage Custom Sites
  [L] List All Sites

Bulk Import File: custom_sites.inf

File Format:

# AUTARCH Adult Site Scanner - Bulk Import File
# Add one domain per line (without http:// or https://)
# Lines starting with # are comments

example.com
another-site.net
subdomain.site.org

Workflow:

User adds domains to custom_sites.inf (one per line)
Run Bulk Import [B]
System reads domains from file
Skips already-added domains
Prompts for test username (e.g., "admin", "test")
Auto-detects URL pattern for each domain
Adds working sites to custom sites
Reports failed domains
Option to clear import file when done

Example Session:

Found 5 domain(s) in custom_sites.inf:
  - site1.com
  - site2.net
  - site3.org
  ...

Will scan 5 new domain(s)

Test username: admin

Scanning 5 domains...

[1/5] Scanning site1.com...
  [+] Added Site1: /user/{}
[2/5] Scanning site2.net...
  [+] Added Site2: /profile/{}
[3/5] Scanning site3.org...
  [X] No pattern found
...

Successfully added 4 site(s)

Failed to detect patterns for 1 domain(s):
  - site3.org
Try adding these manually with [A] or [D]

Clear import file? (y/n): y
[+] Import file cleared

Features:

Creates template file if not exists
Skips domains already in custom sites
Uses first working pattern found (efficient)
Reports failures for manual follow-up
Option to clear file after import

Files Modified

File	Changes
modules/adultscan.py	Added bulk import (~120 new lines)

New Components

Component	Purpose
`BULK_IMPORT_FILE`	Path to `custom_sites.inf`
`probe_domain()`	Reusable pattern detection helper
`bulk_import()`	Bulk import wizard

File Locations

File	Purpose
`custom_sites.inf`	Domains to import (user editable)
`custom_adultsites.json`	Saved custom sites (auto-managed)

Session 1 (Continued) - CLI System & Documentation

User Request

give me an overview of what we have done so far and create in depth technical user guide with all the cli commands and arguments and add a detailed --help argument

Work Completed

14. Comprehensive CLI System

autarch.py - Complete rewrite with argparse CLI

New CLI Options:

Usage: python autarch.py [OPTIONS] [COMMAND]

Options:
  -h, --help              Show detailed help message
  -v, --version           Show version information
  -c, --config FILE       Use alternate config file
  -m, --module NAME       Run a specific module directly
  -l, --list              List all available modules
  --list-category CAT     List modules in specific category
  --show-config           Display current configuration
  --setup                 Force run setup wizard
  --no-banner             Suppress ASCII banner
  -q, --quiet             Minimal output mode
  --verbose               Enable verbose output

Commands:
  chat                    Start interactive LLM chat
  agent                   Start autonomous agent
  osint <username>        Quick OSINT username scan
  scan <target>           Quick port scan
  analyze                 Run analyze module

Examples:

# Show help
python autarch.py --help

# Run specific module
python autarch.py -m adultscan
python autarch.py -m recon

# List modules
python autarch.py --list
python autarch.py --list-category osint

# Quick OSINT
python autarch.py osint targetuser

# Show config
python autarch.py --show-config

# Re-run setup
python autarch.py --setup

Help Output Features:

Detailed epilog with categories, modules, examples
Configuration reference
File locations
Color-coded output

15. Technical User Guide

GUIDE.md - Comprehensive documentation created

Contents:

Project Overview
Project Structure
Installation & Setup
Command Line Interface (all options)
Main Menu Navigation
Module Reference (all modules)
Configuration Reference
Creating Custom Modules
Agent Tools Reference
Troubleshooting
Security Notice

Files Created/Modified

File	Changes
autarch.py	Complete CLI rewrite (~480 lines)
GUIDE.md	New comprehensive guide (~600 lines)

New CLI Functions

Function	Purpose
`create_parser()`	Build argparse parser with all options
`get_epilog()`	Generate detailed help epilog
`show_version()`	Display version info
`show_config()`	Display current config
`list_modules()`	List available modules
`run_module()`	Run module directly
`quick_osint()`	Quick OSINT scan
`quick_scan()`	Quick port scan

Project Summary

Total Files Created: 19 Python files + 3 documentation files

Core Framework (core/):

File	Lines	Purpose
banner.py	~42	ASCII banner, colors
config.py	~143	Configuration handler
llm.py	~280	LLM wrapper
agent.py	~320	Autonomous agent
tools.py	~500	Tool registry
menu.py	~480	Main menu system
msf.py	~380	Metasploit RPC

Modules (modules/):

File	Lines	Category	Purpose
setup.py	~234	-	Setup wizard
chat.py	~250	core	LLM chat
agent.py	~175	core	Agent interface
msf.py	~420	offense	MSF interface
defender.py	~280	defense	System hardening
counter.py	~350	counter	Threat detection
analyze.py	~320	analyze	Forensics
recon.py	~590	osint	OSINT recon
adultscan.py	~650	osint	Adult site scanner
simulate.py	~310	simulate	Attack simulation

Documentation:

File	Purpose
DEVLOG.md	Development history
GUIDE.md	User guide
README.md	(not created, can add)

Configuration Files:

File	Purpose
autarch_settings.conf	Main config
custom_adultsites.json	Custom sites storage
custom_sites.inf	Bulk import file

Feature Summary

LLM Integration:

llama-cpp-python wrapper
ChatML format prompts
Streaming support
Configurable parameters

Agent System:

Autonomous task execution
Tool registry (12+ tools)
THOUGHT/ACTION/PARAMS format
User interaction support

Metasploit Integration:

RPC client
Module search/execution
Session management
Console commands

OSINT Capabilities:

Email lookup/permutation
Username search (17+ platforms)
Phone number analysis
Domain/IP reconnaissance
Adult site scanner (50+ sites)
Custom site management
Auto-pattern detection
Bulk import

Security Tools:

System hardening audit
Threat detection
File forensics
Attack simulation
Payload generation

CLI Features:

Full argparse integration
Module direct execution
Quick commands
Configuration display
Verbose/quiet modes

Session 2 - 2026-01-15

User Request

lets add a new option to defensive 'My System'. When user selects this option it prompts the user to run the full system audit, save the info to system.inf, and scores your security. Then it shows the results as options. when the selects one, it offers to use the llm to auto fix the issue or shows you the options on how to do it manually. Lets also add a CVE database, not just for the system audit, but for other modules as well. For my system menu, have the application detect the system the application is running on and download the database. have the app use https://nvd.nist.gov/developers/vulnerabilities to get the info.

Work Completed

1. CVE Database System (core/cve.py)

CVEDatabase Class - Full NVD API Integration:

Uses NIST NVD REST API v2.0 (https://services.nvd.nist.gov/rest/json/cves/2.0)
Automatic OS detection with CPE mapping
Supports 15+ operating systems:
- Ubuntu, Debian, Fedora, CentOS, RHEL
- Rocky Linux, Alma Linux, Arch, openSUSE, SUSE
- Kali, Linux Mint, Windows, macOS

Key Methods:

Method	Purpose
`_detect_system()`	Auto-detect OS type, version, kernel
`search_cves()`	Search NVD by keyword, CPE, severity
`get_cve_details()`	Get detailed CVE information
`get_system_cves()`	Get CVEs for detected system
`get_software_cves()`	Search CVEs for specific software
`get_installed_packages()`	List system packages (dpkg/rpm/pacman)

Features:

Local JSON cache (24-hour expiry)
API key support for higher rate limits
CVSS v2/v3 score parsing
CPE-based vulnerability matching
Severity filtering (LOW/MEDIUM/HIGH/CRITICAL)
Progress callbacks for UI integration

OS to CPE Mapping:

OS_CPE_MAP = {
    'ubuntu': 'cpe:2.3:o:canonical:ubuntu_linux',
    'debian': 'cpe:2.3:o:debian:debian_linux',
    'fedora': 'cpe:2.3:o:fedoraproject:fedora',
    'rhel': 'cpe:2.3:o:redhat:enterprise_linux',
    'windows': 'cpe:2.3:o:microsoft:windows',
    'macos': 'cpe:2.3:o:apple:macos',
    # ... and more
}

2. My System Module (modules/mysystem.py)

Comprehensive System Audit with CVE Detection & Auto-Fix

Menu Structure:

  My System - Security Audit
  ──────────────────────────────────────────────────
  Detected: ubuntu 22.04
  Kernel: 5.10.0-1012-rockchip
  Last Score: 75/100
  Open Issues: 5

  [1] Run Full System Audit
  [2] Run Audit (Skip CVE Check)

  [3] View Issues (X found)
  [4] View CVE Report

  [5] Search CVE Database
  [6] Check Software for CVEs

  [0] Back to Main Menu

Security Checks Performed:

Check	Description	Severity Impact
Firewall	iptables/ufw/firewalld status	HIGH if missing
SSH Config	Root login, password auth, protocol	HIGH-CRITICAL
Open Ports	15 high-risk ports detection	MEDIUM-CRITICAL
Users	UID 0 accounts, empty passwords	CRITICAL
Permissions	Critical file modes (/etc/shadow, etc.)	MEDIUM
Services	Dangerous services (telnet, rsh, etc.)	HIGH
Updates	Pending package updates	MEDIUM
Fail2Ban	Brute-force protection status	LOW-MEDIUM
Antivirus	ClamAV or other AV detection	LOW
CVEs	System-specific vulnerabilities	HIGH-CRITICAL

Issue Tracking:

SecurityIssue class with severity levels
Automatic security score calculation (0-100)
Score penalties: CRITICAL=-20, HIGH=-15, MEDIUM=-10, LOW=-5
Persists to system.inf JSON file

Issue Remediation Options:

  Issue Details
  ──────────────────────────────────────────────────
  Name: SSH Root Login Enabled
  Severity: HIGH
  Category: ssh

  Description:
    Root login via SSH is not disabled

  Manual Fix Instructions:
    Edit /etc/ssh/sshd_config:
      PermitRootLogin no
    Then restart: sudo systemctl restart sshd

  Auto-Fix Command:
    sudo sed -i 's/^#*PermitRootLogin.*/PermitRootLogin no/' /etc/ssh/sshd_config && sudo systemctl restart sshd

  [1] Auto-Fix with LLM
  [2] Apply Manual Fix
  [3] Mark as Ignored
  [0] Back

LLM Auto-Fix Feature:

Consults LLM for fix recommendations
Provides risk explanation
Generates context-aware fix commands
User confirmation before execution
Streaming response display

CVE Features:

CVE report with severity breakdown
Interactive CVE search
Software-specific CVE lookup
Detailed CVE view with CVSS scores

3. Defender Module Update

modules/defender.py - Added "My System" option:

  System Defender
  ──────────────────────────────────────────────────

  [M] My System - Full audit with CVE detection & auto-fix

  [1] Quick Security Audit
  [2] Firewall Check
  ...

Updated Project Structure

dh_framework/
├── autarch.py
├── autarch_settings.conf
├── system.inf                 # Audit results (NEW)
├── DEVLOG.md
├── GUIDE.md
├── data/
│   └── cve/
│       └── cve_cache.json     # CVE cache (NEW)
├── core/
│   ├── __init__.py
│   ├── agent.py
│   ├── banner.py
│   ├── config.py
│   ├── cve.py                 # CVE database (NEW)
│   ├── llm.py
│   ├── menu.py
│   ├── msf.py
│   └── tools.py
└── modules/
    ├── __init__.py
    ├── adultscan.py
    ├── agent.py
    ├── analyze.py
    ├── chat.py
    ├── counter.py
    ├── defender.py            # Updated with My System
    ├── msf.py
    ├── mysystem.py            # My System module (NEW)
    ├── recon.py
    ├── setup.py
    └── simulate.py

Files Added/Modified

File	Lines	Purpose
core/cve.py	~500	CVE database with NVD API
modules/mysystem.py	~680	My System audit module
modules/defender.py	+10	Added My System menu option

Configuration (autarch_settings.conf)

New optional section for NVD API:

[nvd]
api_key =   ; Optional - for higher rate limits

Storage Files

File	Format	Purpose
system.inf	JSON	Audit results, issues, scores
data/cve/cve.db	SQLite	CVE database
custom_apis.json	JSON	Custom API configurations

Notes

NVD API has rate limits: 5 requests/30s without key, 50 requests/30s with key
Request API key at: https://nvd.nist.gov/developers/request-an-api-key
SQLite database enables fast offline CVE queries
LLM auto-fix requires loaded model
Security score is cumulative based on issue severity

Session 2 (Continued) - SQLite CVE Database & Settings Menus

User Request

what database format did you use for the database... yes lets use sqlite in the settings menu, lets add a CVE menu. Lets also add a menu for users to add custom api's and a menu for this applications api (not implemented yet)

Work Completed

1. SQLite CVE Database (core/cve.py rewrite)

Replaced JSON caching with full SQLite database:

Database Schema:

-- Main CVE table
CREATE TABLE cves (
    id INTEGER PRIMARY KEY,
    cve_id TEXT UNIQUE NOT NULL,
    description TEXT,
    published TEXT,
    modified TEXT,
    cvss_v3_score REAL,
    cvss_v3_severity TEXT,
    cvss_v3_vector TEXT,
    cvss_v2_score REAL,
    cvss_v2_severity TEXT,
    cvss_v2_vector TEXT
);

-- Affected products (CPE)
CREATE TABLE cve_cpes (
    cve_id TEXT,
    cpe_criteria TEXT,
    vulnerable INTEGER,
    version_start TEXT,
    version_end TEXT
);

-- References
CREATE TABLE cve_references (
    cve_id TEXT,
    url TEXT,
    source TEXT
);

-- Weaknesses (CWE)
CREATE TABLE cve_weaknesses (
    cve_id TEXT,
    cwe_id TEXT
);

-- Metadata
CREATE TABLE metadata (
    key TEXT PRIMARY KEY,
    value TEXT
);

Key Methods:

Method	Purpose
`sync_database()`	Download CVEs from NVD API
`sync_recent()`	Quick sync (last 7 days)
`search_cves()`	Local database search
`get_cve()`	Get detailed CVE info
`get_system_cves()`	CVEs for detected OS
`get_software_cves()`	CVEs for specific software
`fetch_cve_online()`	Online fallback for single CVE
`search_online()`	Online search fallback

Features:

Thread-safe SQLite connections
Indexed columns for fast queries
Batch processing with progress display
Rate limiting (respects NVD limits)
Online fallback when database empty

New Settings Menu Structure:

  Settings
  ──────────────────────────────────────────────────

  [1] LLM Settings
  [2] Metasploit Settings
  [3] CVE Database Settings       <- NEW
  [4] Custom APIs                 <- NEW
  [5] AUTARCH API                 <- NEW

  [6] View All Settings
  [7] Run Setup Wizard

  [0] Back

  CVE Database Settings
  ──────────────────────────────────────────────────

  Database Path: /home/.../data/cve/cve.db
  Database Size: 150.5 MB
  Total CVEs: 245,000
  Last Sync: 2026-01-15

  Detected OS: Ubuntu 22.04.5 LTS
  CPE Prefix: cpe:2.3:o:canonical:ubuntu_linux

  NVD API Key: Configured

  [1] Sync Database (Recent - 120 days)
  [2] Sync Database (Full - all CVEs)
  [3] Set NVD API Key
  [4] Clear Database

  [0] Back

Allows users to add and manage external API integrations:

  Custom APIs
  ──────────────────────────────────────────────────

  Configured APIs:
    [1] VirusTotal - Active
        https://www.virustotal.com/api/v3/...
    [2] Shodan - Active
        https://api.shodan.io/...

  [A] Add API
  [E] Edit API
  [D] Delete API
  [T] Toggle API

  [0] Back

API Configuration Fields:

Name
Base URL
API Key
Description
Type (REST, GraphQL, SOAP, Other)
Enabled/Disabled status

Storage: custom_apis.json

Placeholder for future REST API implementation:

  AUTARCH API
  ──────────────────────────────────────────────────

  Status: Disabled
  Port: 8080
  API Key: Not set

  [!] API functionality coming in future version

  [1] Configure API Settings
  [2] Generate API Key
  [3] View API Documentation

  [0] Back

Planned Endpoints:

GET /api/v1/status - Framework status
GET /api/v1/modules - List modules
POST /api/v1/scan - Run security scan
GET /api/v1/cve/search - Search CVE database
POST /api/v1/agent/task - Submit agent task

Updated My System Module

Added CVE database sync options:

  My System - Security Audit
  ──────────────────────────────────────────────────
  Detected: ubuntu 22.04
  CVE Database: 245,000 CVEs (150.5 MB)
  Last Sync: 2026-01-15

  [1] Run Full System Audit
  [2] Run Audit (Skip CVE Check)

  [7] Sync CVE Database (Recent)   <- NEW
  [8] Sync CVE Database (Full)     <- NEW
  [9] CVE Database Info            <- NEW

  [0] Back

Files Modified

File	Changes
core/cve.py	Complete rewrite - SQLite database (~870 lines)
core/menu.py	Added CVE, Custom APIs, AUTARCH API menus (~300 new lines)
modules/mysystem.py	Updated for SQLite, added sync options (~100 lines changed)

New Files

File	Purpose
data/cve/cve.db	SQLite CVE database
custom_apis.json	Custom API configurations

Database Sync Estimates

Sync Type	CVEs	Time (no key)	Time (with key)	Size
Recent (120 days)	~5,000	10-15 min	2-3 min	~5 MB
Full (since 1999)	~245,000	4-6 hours	30-60 min	~150-300 MB

Notes

SQLite file located at data/cve/cve.db
Get NVD API key for faster syncs: https://nvd.nist.gov/developers/request-an-api-key
Database supports offline CVE lookups after initial sync
Custom APIs stored in custom_apis.json in framework root
AUTARCH API is placeholder - implementation in future version

Session 2 (Continued) - Sites Database Expansion

User Request

now lets add more sites. Start crawling and scraping, do not exclude any kind of site. If you can create an account and post things add it. both nsfw and sfw

Work Completed

1. New Source Added: reveal-my-name

Added osint-liar/reveal-my-name as a new source (extended WhatsMyName fork with 2,140+ sites):

URL: https://raw.githubusercontent.com/osint-liar/reveal-my-name/main/wmn-data.json
Contains 628 parseable sites with improved detection patterns
Handles XXXPORNXXX category for NSFW detection

Parser Added: _parse_reveal_my_name() in core/sites_db.py

2. XenForo/vBulletin Forums Added

Added 43 major forums from XenForo's large forums list with multiple URL patterns:

Posts	Forums Added
100M+	IGN Boards
50-99M	Disboards, Christian Forums, BigFooty
20-49M	Sherdog, HFBoards, PurseForum, SpaceBattles, ADV Rider, Grasscity, etc.
10-19M	Paradox, BladeForums, Smashboards, RedCafe, TalkBass, TheColi, Se7enSins, etc.

Each forum added with both XenForo (/members/{}.html) and vBulletin (/member.php?username={}) patterns.

3. Adult/NSFW Sites Added

Cam Sites:

Chaturbate, StripChat, CamSoda, BongaCams, LiveJasmin, Cam4, MyFreeCams
JerkMate, LivePrivates, Flirt4Free, Streamate

Fan/Creator Platforms:

OnlyFans, Fansly, JustForFans, Fanvue, ManyVids
LoyalFans, FanCentro, PocketStars, Unlockd, Alua, AdmireMe VIP

Tube Sites:

Pornhub Models, xHamster Models, XVideos Models, ModelHub

Adult Social/Dating:

FetLife, CollarSpace, SwingLifeStyle, Adult Friend Finder, Ashley Madison

Gaming Adult:

F95zone, LoversLab, ULMF

Hentai/Anime:

Hentai Foundry, Fakku, Gelbooru, Danbooru, Sankaku Complex

Furry:

Fur Affinity, e621, SoFurry, Inkbunny

4. Mainstream Sites Added

Social/Messaging:

Discord, Telegram, Mastodon, Threads, Bluesky, Cohost, Nostr, Matrix

Dating:

Tinder, Bumble, Hinge, OkCupid, Badoo, Grindr, Scruff, HER

Crypto/Finance:

CoinMarketCap, CoinGecko, OpenSea, Rarible, Foundation, Mirror, Farcaster

Streaming:

Twitch, Kick, TikTok, Trovo, DLive, Rumble, Odysee

Creative:

ArtStation, Dribbble, Behance, DeviantArt, SoundCloud, Bandcamp

Shopping:

Etsy, eBay, Depop, Poshmark, Grailed, Fiverr, Upwork

Blogging:

Medium, Substack, Dev.to, Hashnode, Ghost

Database Statistics

Metric	Before	After
Total Sites	3,481	3,751
NSFW Sites	618	661
Custom Sites	400	672
Sources	7	8

Sites by Source:

Source	Count
maigret	1,682
custom	672
social_analyzer	469
reveal_my_name	312
sherlock	177
cupidcr4wl	142
whatsmyname	137
detectdee	126
nexfil	34

Sites by Category:

Category	Count
other	1,922
adult	647
social	258
forum	245
gaming	166
tech	152
news	69
finance	68
dating	38
art	37

Files Modified

File	Changes
core/sites_db.py	Added reveal_my_name source URL, parser, and import support
data/sites/reveal_my_name.json	Downloaded 7,076 lines (628 parseable sites)
data/sites/sites.db	Updated with 270+ new sites

Notes

reveal_my_name prioritized below maigret/sherlock but above whatsmyname
XXXPORNXXX category auto-mapped to 'adult' with nsfw=1
XenForo forums added with both XenForo and vBulletin URL patterns
Many adult sites use simple status code detection (200 = exists)

Session 2 (Continued) - Snoop Project Integration

User Request

i have it installed here... and the full database is BDfull

Work Completed

1. Snoop Database Decoding

The Snoop Project stores its database in a base32-encoded, reversed format. Decoded the full database:

Decoding Process:

# 1. Read base32 encoded file
db = file.read()
# 2. Decode base32
db_bytes = base64.b32decode(db)
# 3. Reverse bytes
db_bytes = db_bytes[::-1]
# 4. Decode UTF-8
content = db_bytes.decode('utf-8', errors='replace')
# 5. Reverse string
content = content[::-1]
# 6. Parse JSON
data = json.loads(content)

Results:

BDfull: 5,366 sites decoded
Saved to: /home/snake/dh_framework/data/sites/snoop_full.json (1.95 MB)

2. Snoop Parser Added

Added _parse_snoop() method to core/sites_db.py:

Snoop Data Structure:

{
  "SiteName": {
    "country": "🇺🇸",           # Emoji flag
    "country_klas": "US",      # Country code
    "errorType": "status_code", # Detection method
    "url": "https://site.com/user/{}", # URL template
    "urlMain": "https://site.com/",    # Main URL
    "usernameON": "adam",      # Test username
    "errorMsg": "Not found",   # Error message
    "bad_site": ""             # Problem indicator
  }
}

Parser Features:

Maps errorType to detection method (status_code → status, message → content)
Extracts error patterns from errorMsg/errorMsg2
Handles encoding issues in key names

Updated Database Statistics

Metric	Before	After
Total Sites	3,751	8,315
NSFW Sites	661	654
Sources	8	9

Sites by Source:

Source	Count
snoop	4,641
maigret	1,727
custom	604
social_analyzer	440
reveal_my_name	308
sherlock	169
cupidcr4wl	145
whatsmyname	134
detectdee	122
nexfil	25

Files Modified/Added

File	Changes
core/sites_db.py	Added snoop source, `_parse_snoop()` method, updated priorities
data/sites/snoop_full.json	5,366 sites (1.95 MB)
data/sites/sites.db	Updated with 8,315 total sites

Backup Created

Path: /home/snake/backups/dh_framework_backup_20260115_044001.tar.gz
Size: 1.5 MB

Notes

Snoop prioritized between maigret (highest) and sherlock
Database now contains 8,315 sites for username enumeration (more than doubled)
Many Snoop sites are Russian/Eastern European forums
Snoop source stored locally (base32 encoded file required)

Session 2 (Continued) - Snoop Decoder Module

User Request

now create a module to decrypt snoop databases and add it to OSINT menu

Work Completed

1. Created Snoop Decoder Module

modules/snoop_decoder.py - Full GUI module for decoding Snoop databases

Features:

Decode any Snoop database file (BDdemo, BDfull, custom)
View decoded data with statistics
Import decoded sites to AUTARCH database
Quick import from known locations
View current sites database stats

Menu Structure:

  Snoop Database Decoder
  ──────────────────────────────────────────────────

  [1] Decode Snoop Database File
  [2] Decode & Import to AUTARCH
  [3] View Current Sites Database Stats

  [4] Quick Import (BDfull from snoop-master)
  [5] Quick Import (BDdemo from snoop-master)

  [0] Back to OSINT Menu

SnoopDecoder Class Methods:

Method	Purpose
`decode_database()`	Decode base32 encoded database file
`save_decoded()`	Save decoded data to JSON
`import_to_database()`	Import sites to AUTARCH SQLite DB
`show_sample()`	Display sample sites from decoded data
`get_stats()`	Get statistics about decoded database

modules/recon.py - Updated with new tools section

New Menu Options:

  Tools
    [S] Snoop Database Decoder
    [D] Sites Database Stats

New Methods Added:

Method	Purpose
`run_snoop_decoder()`	Launch Snoop Decoder module
`show_sites_db_stats()`	Display sites DB statistics with visual bars

Files Created/Modified

File	Lines	Purpose
modules/snoop_decoder.py	~320	New Snoop decoder module
modules/recon.py	+40	Added tools section and handlers

Module Metadata

NAME = "Snoop Decoder"
DESCRIPTION = "Decrypt and import Snoop Project databases"
AUTHOR = "darkHal Security Group"
VERSION = "1.0"
CATEGORY = "osint"

Usage

From AUTARCH main menu:

Select OSINT (option 5)
Press S for Snoop Database Decoder
Choose decode option (1-5)
Follow prompts to decode/import

Or run directly:

python autarch.py -m snoop_decoder

Session - 2026-01-15 (Continued) - Snoop Features Integration

User Request

now lets add some snoops features to our app

Work Completed

Added several Snoop-inspired features to AUTARCH, all translated to English and branded for AUTARCH:

1. GEO IP/Domain Lookup Module (`modules/geoip.py`)

Full geolocation lookup for IPs, domains, and URLs. Features:

Single IP/domain/URL lookup
Your own IP lookup
Bulk lookup from file
Uses multiple API backends (ipwho.is, ipinfo.io)
DNS resolution with IPv4/IPv6
Map links (OpenStreetMap, Google Maps)

NAME = "GEO IP Lookup"
DESCRIPTION = "Get geolocation for IPs, domains, and URLs"
AUTHOR = "darkHal Security Group"
VERSION = "1.0"
CATEGORY = "osint"

2. Yandex OSINT Module (`modules/yandex_osint.py`)

Gather intelligence from Yandex user accounts. Features:

Lookup by login/email
Extract user from Yandex.Disk public links
Lookup by public ID (26-char hash)
Returns: name, email, avatar, and profile links for:
- Yandex Reviews
- Yandex Market
- Yandex Music
- Yandex Dzen
- Yandex Q&A

NAME = "Yandex OSINT"
DESCRIPTION = "Gather intel from Yandex user accounts"
AUTHOR = "darkHal Security Group"
VERSION = "1.0"
CATEGORY = "osint"

3. Network Test Module (`modules/nettest.py`)

Network connectivity and speed testing. Features:

Connectivity test (ping multiple sites)
Full speed test (download/upload/ping)
DNS resolution test
Run all tests option
Uses speedtest-cli library (optional)

NAME = "Network Test"
DESCRIPTION = "Test network speed and connectivity"
AUTHOR = "darkHal Security Group"
VERSION = "1.0"
CATEGORY = "utility"

4. HTML Report Generator (`core/report_generator.py`)

Generate professional HTML reports for scan results. Features:

Dark theme with AUTARCH branding
Username scan reports with:
- Stats overview
- Confidence scoring visualization
- Category breakdown
- Restricted access section
GEO IP bulk lookup reports
Responsive table design

New menu structure in modules/recon.py:

  Tools
    [G] GEO IP/Domain Lookup      <- NEW
    [Y] Yandex OSINT              <- NEW
    [N] Network Test              <- NEW
    [S] Snoop Database Decoder
    [D] Sites Database Stats

6. Username Scanner Improvements

Added scan time tracking
HTML report generation option
Save options: [1] JSON, [2] HTML, [3] Both, [n] No

Files Created

File	Lines	Purpose
modules/geoip.py	~350	GEO IP/Domain lookup
modules/yandex_osint.py	~280	Yandex user OSINT
modules/nettest.py	~300	Network speed/connectivity tests
core/report_generator.py	~350	HTML report generation

Files Modified

File	Changes
modules/recon.py	Added new modules to menu, HTML report support
(version bumped to 2.2)

Directory Structure Update

dh_framework/
├── core/
│   ├── report_generator.py     # NEW - HTML reports
│   └── ...
├── modules/
│   ├── geoip.py                # NEW - GEO IP lookup
│   ├── yandex_osint.py         # NEW - Yandex OSINT
│   ├── nettest.py              # NEW - Network testing
│   └── ...
└── results/
    └── reports/                # NEW - HTML report output

Usage Examples

GEO IP Lookup:

OSINT Menu > [G] GEO IP/Domain Lookup
> Enter: 8.8.8.8
> Shows: Country, Region, City, ISP, Map links

Yandex OSINT:

OSINT Menu > [Y] Yandex OSINT
> Enter Yandex login: username
> Shows: Name, Email, Avatar, Profile links

Network Test:

OSINT Menu > [N] Network Test
> [1] Test Connectivity
> [2] Full Speed Test
> [3] Test DNS
> [4] Run All Tests

Username Scan with HTML Report:

OSINT Menu > [3] Username Lookup
> Enter username: target_user
> Scan completes...
> Save results? [2] HTML
> Saved HTML report to results/reports/target_user_20260115_050000.html

Notes

All modules are in English (translated from Russian Snoop Project)
All modules are branded as AUTARCH/darkHal Security Group
Modules follow AUTARCH coding conventions
HTML reports use a dark theme matching the terminal aesthetic

Session 3 - OSINT Improvements

Overview

This session focused on improving the OSINT username scanning functionality:

Adding configurable thread settings
Fixing malformed site names
Improving false positive detection
Cleaning up garbage sites from the database

Part 1: OSINT Thread Settings

User Request

we need to add a threads setting for the OSINT search. For some reason it thinks my device has 50 threads. So lets add a option in the settings menu to adjust max threads and lets have the application default to 8

Work Completed

1. Added OSINT Configuration Section (core/config.py)

New Default Config Section:

'osint': {
    'max_threads': '8',
    'timeout': '8',
    'include_nsfw': 'false',
}

New Method:

def get_osint_settings(self) -> dict:
    """Get all OSINT settings as a dictionary."""
    return {
        'max_threads': self.get_int('osint', 'max_threads', 8),
        'timeout': self.get_int('osint', 'timeout', 8),
        'include_nsfw': self.get_bool('osint', 'include_nsfw', False),
    }

2. Updated OSINT Modules to Use Config

modules/recon.py:

Imports get_config from core.config
Reads thread count from config instead of hardcoded 50
Also uses config for timeout and NSFW settings

def __init__(self):
    self.config = get_config()
    osint_settings = self.config.get_osint_settings()
    self.scan_config = {
        'max_sites': 200,
        'include_nsfw': osint_settings['include_nsfw'],
        'categories': None,
        'timeout': osint_settings['timeout'],
        'threads': osint_settings['max_threads'],  # Was hardcoded to 50
    }

modules/adultscan.py:

Imports get_config from core.config
Uses self.max_threads from config instead of hardcoded 10

def __init__(self):
    self.config = get_config()
    osint_settings = self.config.get_osint_settings()
    self.timeout = osint_settings['timeout']
    self.max_threads = osint_settings['max_threads']  # Was hardcoded to 10

Updated Settings Menu:

  Settings
  ──────────────────────────────────────────────────

  [1] LLM Settings
  [2] Metasploit Settings
  [3] Database Management
  [4] Custom APIs
  [5] AUTARCH API
  [6] OSINT Settings          <- NEW

  [7] View All Settings
  [8] Run Setup Wizard

  [0] Back

OSINT Settings Submenu:

  OSINT Settings
  ──────────────────────────────────────────────────

    Max Threads:    8
    Timeout:        8 seconds
    Include NSFW:   No

  Thread setting controls parallel requests during
  username scanning. Lower values = slower but safer.

  [1] Set Max Threads
  [2] Set Timeout
  [3] Toggle NSFW Sites

  [0] Back

New Methods Added:

Method	Purpose
`show_osint_settings()`	Display OSINT settings menu
`_set_osint_threads()`	Configure max threads (1-100)
`_set_osint_timeout()`	Configure timeout (1-60 seconds)
`_toggle_osint_nsfw()`	Toggle NSFW site inclusion

4. Updated "View All Settings"

Now includes OSINT configuration in the full settings view:

  OSINT Configuration:
    max_threads         : 8
    timeout             : 8
    include_nsfw        : False

Files Modified

File	Changes
core/config.py	Added `[osint]` section defaults, `get_osint_settings()` method
core/menu.py	Added OSINT Settings menu (option 6), 3 new config methods
modules/recon.py	Import config, use `osint_settings['max_threads']`
modules/adultscan.py	Import config, use `self.max_threads` from config

Configuration File Format

New section in autarch_settings.conf:

[osint]
max_threads = 8
timeout = 8
include_nsfw = false

Usage

To adjust OSINT thread count:

Main Menu → Settings (99) → OSINT Settings (6) → Set Max Threads (1)

Recommended values:

Low-end devices: 4-8 threads
Mid-range devices: 8-16 threads
High-end devices: 16-32 threads

Notes

Default changed from 50 to 8 threads for safer scanning
Setting persists in autarch_settings.conf
Both recon.py and adultscan.py now use the same config
Timeout and NSFW toggle also configurable from the same menu

Part 2: Username Scan Improvements

User Request:

we need improve the false positive detection on the username scan, as well as scan sites alphabetically. It also appears we have a naming issue. we have lots of sites that just say forum_name instead of the name of the sites. Also there are missing sites. Did you filter and remove sites like imgsrc.ru when you imported data?

Investigation Results

Malformed Names: Found 3,409 sites with bad names:
- {username}.domain style names (placeholder not replaced)
- Forum_sitename patterns
- site_vb1, site_xf, site_phpbb duplicates (forum software variants)
imgsrc.ru: NOT filtered - exists in database (2 entries)
Sites ordered by rank, not alphabetically

Work Completed

1. Database Cleanup (core/sites_db.py)

Added cleanup_names() method:

Fixes {username} style names by extracting from URL domain
Fixes Forum_name patterns by extracting actual name
Removes duplicate forum software variants (_vb1, _xf, _phpbb, etc)
Merges renamed entries if name already exists

Cleanup Results:

Renamed: 3,171
Merged:  84
Deleted: 407
Total removed: 3,662 malformed entries
Sites remaining: 7,824

2. Alphabetical Sorting (core/sites_db.py)

Updated get_sites_for_scan() method:

Added sort_alphabetically parameter (default: True)
Sites now scanned A-Z by default instead of by rank
Query excludes malformed names automatically

def get_sites_for_scan(
    self,
    categories: List[str] = None,
    include_nsfw: bool = False,
    max_sites: int = 500,
    sort_alphabetically: bool = True  # NEW
) -> List[Dict]:

SQL Filtering:

AND name NOT LIKE '{%'
AND name NOT LIKE '%_vb1' AND name NOT LIKE '%_vb2'
AND name NOT LIKE '%_xf' AND name NOT LIKE '%_phpbb'
AND name NOT LIKE '%_mybb' AND name NOT LIKE '%_smf'
AND name NOT LIKE '%_ipb' AND name NOT LIKE '%_generic'

3. Improved False Positive Detection (modules/recon.py)

Expanded NOT_FOUND_PATTERNS (30 patterns):

Registration prompts ("this username is available")
Soft 404 indicators ("oops", "sorry")
Suspension/ban messages
Generic error page patterns
Title tag checks for 404/error

Expanded FOUND_PATTERNS (23 patterns):

Account age/dates
Activity statistics
Activity timestamps
Profile content indicators
Verification badges
Cam/streaming site patterns
Social profile patterns
E-commerce/creator patterns

Added FALSE_POSITIVE_URLS list:

FALSE_POSITIVE_URLS = [
    '/login', '/signin', '/signup', '/register', '/join',
    '/404', '/error', '/not-found', '/notfound',
    '/search', '/home', '/index', '/welcome',
]

Improved Detection Logic:

Username variation checking (underscores, hyphens, dots)
Better handling of short usernames (extra validation required)
Short page content checks for generic indicators
API/JSON endpoint confidence reduction
Search query parameter detection
Confidence capping at 100%
Higher minimum threshold (45% vs 40%)

Files Modified

File	Changes
core/sites_db.py	`cleanup_names()` method, `sort_alphabetically` param, malformed name filtering
modules/recon.py	Expanded patterns, FALSE_POSITIVE_URLS, improved detection logic

Site Count After Cleanup

Metric	Count
Total sites	7,824
Malformed names	0
imgsrc.ru entries	2 (not filtered)

Notes

No sites were filtered during import - imgsrc.ru and other adult sites are present
Forum software variants were removed as duplicates (one entry per forum is sufficient)
Alphabetical sorting makes progress easier to track during long scans
False positive detection now more robust with 30+ NOT_FOUND patterns

Part 3: Other Category Cleanup

User Request:

its seems like their is alot of garbage sites in the other category

Investigation

Found 6,462 sites (82%) in "other" category with many issues:

Russian forum farms (ucoz, borda, at.ua, clan.su)
Search URLs (not actual profile pages)
Dead/closed sites
Wiki user pages
Invalid domains (google.com, gmail.com)
Duplicate entries

Work Completed

1. Added `cleanup_garbage_sites()` Method

Disables low-quality sites:

# Russian forum farms
ucoz.ru, ucoz.net, ucoz.com, at.ua, borda.ru, clan.su, forum24.ru, mybb.ru, do.am

# Search URLs (not profile pages)
search.php?author=, /search?, action=search, memberlist.php?mode=viewprofile

# uCoz profile pattern
/index/8-0-

Deletes garbage:

# Dead sites
CLOSEDEAD, CLOSED, __DEAD, _DEAD

# Duplicate markers
__2, __3

2. Added `auto_categorize()` Method

Auto-categorizes sites based on name/URL patterns:

tech: github, stackoverflow, hackerone, etc.
gaming: twitch, steam, xbox, playstation, etc.
art: 500px, flickr, deviantart, etc.
forum: sites with forum/forums in URL
adult: pornhub, onlyfans, chaturbate, etc.
social: mastodon, minds, mewe, etc.
And more...

3. Added `remove_duplicates()` Method

Removes sites with identical URL templates.

4. Additional Cleanup

Disabled wiki user pages (/wiki/User:)
Disabled archive.org wayback URLs
Deleted invalid domains (google.com, gmail.com)
Disabled more search URL patterns

Results

Metric	Before	After	Change
Total sites	7,824	7,119	-705
Enabled sites	7,824	4,786	-3,038
"other" category	6,462 (82%)	2,011 (42%)	-4,451
Disabled sites	0	2,333	+2,333

Sites by Category (Enabled):

other        2,011
forum        1,284
social         277
adult          243
tech           240
gaming         170
art             95
news            91
video           82
finance         80
music           63
professional    50
shopping        46
dating          45
hobby            8
images           1

New Methods in sites_db.py

Method	Purpose
`cleanup_garbage_sites()`	Disable Russian forums, search URLs, dead sites
`auto_categorize()`	Auto-categorize "other" sites by patterns
`remove_duplicates()`	Remove duplicate URL templates
`get_disabled_count()`	Get count of disabled sites
`enable_all_sites()`	Re-enable all disabled sites

Notes

Sites are disabled, not deleted (can be re-enabled)
"other" category now contains legitimate misc sites
Quality over quantity - 4,786 enabled sites vs 7,824 total
Use db.enable_all_sites() to restore all sites if needed

Session 3 Summary

Files Modified:

File	Changes
`core/config.py`	Added `[osint]` section, `get_osint_settings()`
`core/menu.py`	Added OSINT Settings menu (option 6)
`core/sites_db.py`	Added `cleanup_garbage_sites()`, `auto_categorize()`, `remove_duplicates()`, `cleanup_names()`, alphabetical sorting
`modules/recon.py`	Expanded detection patterns, improved confidence logic
`modules/adultscan.py`	Added config support for threads

Database Changes:

Metric	Original	Final
Total sites	8,315+	7,119
Enabled sites	8,315+	4,786
"other" category	82%	42%
Malformed names	3,409	0

Key Improvements:

Configurable OSINT thread count (default: 8)
Sites scanned alphabetically for easier progress tracking
30+ NOT_FOUND patterns for false positive detection
23+ FOUND patterns for profile validation
Auto-categorization of sites
Garbage site filtering (Russian forum farms, search URLs, wiki pages)

User Request

the username search still needs some tweaks. take a look at how social analyzer does it /home/snake/Downloads/OSINT/social-analyzer-main.zip and make ours more like that

Examined the social-analyzer codebase and identified key differences:

Detection System: Uses return: true/false pattern
- return: false + string found = user does NOT exist
- return: true + string found = user EXISTS
Rate Calculation: rate = (detections_passed / detections_total) * 100
Status Categories:
- good: 100% rate
- maybe: 50-100% rate
- bad: <50% rate
WAF/Captcha Detection: Filters Cloudflare and captcha pages
Random Delays: sleep(randint(1, 99) / 100) to avoid rate limiting
Retry Logic: Retries failed sites up to 3 times

Work Completed

1. Rewrote Detection System (modules/recon.py)

New WAF/Captcha Detection:

WAF_PATTERNS = re.compile(
    r'captcha-info|Please enable cookies|Completing the CAPTCHA|'
    r'checking your browser|just a moment|ddos protection|'
    r'access denied|blocked|security check|verify you are human',
    re.IGNORECASE
)

WAF_TITLE_PATTERNS = re.compile(
    r'not found|blocked|attention required|cloudflare|'
    r'access denied|security check|ddos|captcha',
    re.IGNORECASE
)

New Shared Detections (like social-analyzer):

SHARED_DETECTIONS = {
    'mastodon': [
        {'return': False, 'string': "The page you are looking for isn"},
        {'return': True, 'string': 'profile:username'},
    ],
    'discourse': [...],
    'gitlab': [...],
    'phpbb': [...],
    'xenforo': [...],
    'vbulletin': [...],
}

New Detection Logic:

Check NOT_FOUND_STRINGS (return: false patterns)
Check FOUND_STRINGS (return: true patterns)
Check if username in content/URL
Calculate rate as percentage
Determine status (good/maybe/bad)

2. Updated `_check_site()` Method

Key Changes:

Random delay: time.sleep(randint(5, 50) / 100)
Cloudflare detection via cf-ray header
WAF content pattern matching
Title extraction and analysis
Rate calculation: rate = (passed / total) * 100
Status determination: good (100%), maybe (50-100%), bad (<50%)
Retry logic for 5xx errors and connection failures (up to 2 retries)
Returns filtered status for WAF-blocked pages

Return Format:

{
    'name': site['name'],
    'url': url,
    'category': site.get('category', 'other'),
    'rate': '75.5%',      # Percentage
    'status': 'maybe',    # good/maybe/bad/restricted/filtered
    'title': 'Page Title',
    'is_tracker': False,
    'found': 5,           # Detections passed
    'total': 8,           # Total detections
}

3. Updated Display Logic

Real-time Output:

  [+] SiteName                https://site.com/user [100.0%]  (good - green)
  [?] OtherSite               https://other.com/user [65.5%]  (maybe - yellow)

Summary Categories:

Results Breakdown (social-analyzer style):
  Detected (good):     15
  Unknown (maybe):     8
  Bad (low rate):      3
  Restricted (403):    12
  Filtered (WAF):      5
  Tracker sites:       2

Files Modified

File	Changes
modules/recon.py	Complete rewrite of detection system, WAF detection, retry logic, rate calculation

Detection Patterns

NOT_FOUND_STRINGS (return: false):

user not found, profile not found, page not found
does not exist, no user, no such user
account has been suspended/deleted/banned
sign up, create an account, register now
there's nothing here, this page is no longer available

FOUND_STRINGS (return: true):

og:title, profile:username, user-profile
member-header, profile-header, user-info
followers, following, posts, joined
member since, last seen, last active
follow, subscribe, message

Feature	Before	After (social-analyzer style)
Detection	Regex patterns	String matching with return true/false
Confidence	0-100 score	Rate percentage
Status	High/Medium/Low	good/maybe/bad
WAF Detection	None	Cloudflare + captcha patterns
Delays	None	Random 50-500ms
Retries	None	Up to 2 retries
Filtering	None	Filtered status for WAF

Notes

Detection system now mirrors social-analyzer's approach
Rate calculation is more accurate than arbitrary confidence scores
WAF detection prevents false positives from Cloudflare pages
Random delays reduce rate limiting issues
Retry logic handles temporary failures
Filtered status clearly marks WAF-blocked results

Session 3 (Continued) - Verbose Output & Blackbird Import

User Request

you turned off username scan verbose results, make sure it shows realtime the site being scanned. Also import the sites from this application /home/snake/blackbird

Work Completed

1. Real-Time Verbose Output (modules/recon.py)

Added real-time progress display during username scanning:

# Show current site being checked (verbose)
print(f"\r{Colors.DIM}  [{checked}/{total_sites}] Checking: {site['name'][:30]:30}{Colors.RESET}", end='', flush=True)

Features:

Shows current progress counter [X/Y]
Displays site name being checked (truncated to 30 chars)
Uses carriage return \r for in-place updates
Dimmed color to distinguish from results
Clears line before printing final results

2. Blackbird Sites Import (core/sites_db.py)

Added import_from_blackbird() method to import sites from the blackbird OSINT tool.

Blackbird Data Format (wmn-data.json):

{
  "SiteName": {
    "main": "https://example.com",
    "uri_check": "https://example.com/user/{account}",
    "e_string": "not found",
    "e_code": 404,
    "m_string": "",
    "m_code": 200,
    "cat": "social"
  }
}

Import Method:

def import_from_blackbird(self, blackbird_path: str = '/home/snake/blackbird', verbose: bool = True) -> Dict[str, int]:
    """Import sites from blackbird application."""
    # Load wmn-data.json
    # Parse each site entry
    # Handle {account} placeholder -> {}
    # Handle name collisions by adding _bb suffix
    # Skip duplicate URLs

Name Collision Handling:

# Check if name already exists - if so, append source suffix
cursor.execute("SELECT id FROM sites WHERE name = ?", (name,))
existing_name = cursor.fetchone()
if existing_name:
    name = f"{name}_bb"  # Add blackbird suffix

Import Results:

First run: 77 new sites added, 573 skipped (URL duplicates)
After name collision fix: 91 additional sites added
Total from blackbird: 168 new sites

Files Modified

File	Changes
modules/recon.py	Added real-time verbose output in `_check_site()` loop
core/sites_db.py	Added `import_from_blackbird()` method

Database Statistics Update

Metric	Before	After
Total sites	7,119	7,287
Sources	9	10
Blackbird sites	0	168

Notes

Blackbird uses {account} as placeholder (converted to {})
Name collisions resolved with _bb suffix
Most blackbird sites already existed in database (from other sources)
Verbose output updates in-place without scrolling

Session 4 - 2026-01-15 (Continued) - Dossier Manager & NSFW Fix

User Request

in the OSINT section, lets add a new menu named Dossier. In dossier the options are Start New and View. What this module does is lets users view saved information from the recon module and lets you associate information such as the results from an email search and username search

Work Completed

1. Dossier Manager Module (modules/dossier.py)

Created a comprehensive OSINT investigation management system:

Menu Structure:

  Dossier Manager
  ──────────────────────────────────────────────────
  Saved dossiers: X

  [1] Start New Dossier
  [2] View Dossiers

  [0] Back

Dossier Features:

Create new dossiers with subject name, notes, and initial identifiers
Store multiple identifier types:
- Emails
- Usernames
- Phone numbers
- Real names
- Aliases
Import search results from JSON files (username scan results)
Manually add profiles
Add investigation notes
View all associated data grouped by category
Export as JSON or text report

Dossier Detail Menu:

  View
    [1] View Identifiers
    [2] View Search Results
    [3] View Profiles
    [4] View Notes

  Add
    [5] Add Identifier
    [6] Import Search Results
    [7] Add Profile Manually
    [8] Add Note

  Manage
    [E] Edit Dossier Info
    [X] Export Dossier
    [D] Delete Dossier

Storage:

Dossiers saved as JSON in dossiers/ directory
Auto-generated unique IDs with timestamps
Supports importing *_profiles.json files from username scans

New Menu Section:

  Dossier
    [R] Dossier Manager

Methods Added:

Method	Purpose
`run_dossier_manager()`	Launch Dossier Manager module

3. Fixed NSFW Adult Site Detection

Issue: Adult sites like Chaturbate weren't appearing in results even with NSFW enabled.

Root Causes Found:

Inconsistent NSFW flags: 97 adult category sites had nsfw=0
Config not used: include_nsfw config setting wasn't being used as default

Fixes Applied:

Database Fix:

UPDATE sites SET nsfw = 1 WHERE category = 'adult'
-- Updated 179 adult category sites

Code Fix (modules/recon.py):

# Before: hardcoded default
include_nsfw = False

# After: uses config setting
osint_settings = self.config.get_osint_settings()
include_nsfw = osint_settings['include_nsfw']

Prompt Updated:

Now shows current config default (y/n)
Press Enter keeps default instead of overriding to 'n'

Files Created

File	Lines	Purpose
modules/dossier.py	~680	Dossier Manager module

Files Modified

File	Changes
modules/recon.py	Added Dossier menu section, `run_dossier_manager()` method, fixed NSFW default
data/sites/sites.db	Fixed nsfw=1 for all adult category sites

New Directory

Path	Purpose
dossiers/	Storage for dossier JSON files

Dossier JSON Structure

{
  "meta": {
    "name": "Investigation Name",
    "subject": "Target identifier",
    "created": "2026-01-15T12:00:00",
    "modified": "2026-01-15T12:00:00",
    "notes": "Initial notes"
  },
  "identifiers": {
    "emails": ["user@example.com"],
    "usernames": ["username1", "username2"],
    "phones": ["+1234567890"],
    "real_names": ["John Doe"],
    "aliases": ["alias1"]
  },
  "results": {
    "email_searches": [],
    "username_searches": [
      {
        "username": "target",
        "date": "2026-01-15T12:00:00",
        "total_checked": 500,
        "found": [...]
      }
    ],
    "phone_searches": []
  },
  "profiles": [
    {
      "name": "SiteName",
      "url": "https://site.com/user",
      "category": "social",
      "status": "good",
      "rate": "100%"
    }
  ],
  "custom_notes": [
    {"date": "2026-01-15T12:00:00", "text": "Investigation note"}
  ]
}

Notes

Dossier Manager allows correlating data from multiple OSINT searches
Import feature automatically adds username to identifiers list
All adult category sites now properly flagged as NSFW
Username scan now respects include_nsfw config setting as default
Dossiers can be exported as JSON (full data) or text (readable report)

Session 4 (Continued) - Site Additions & Adult Site Fixes

User Requests

chaturbate.com should have user url like this https://chaturbate.com/fudnucker/ in the project directory add the sites from pred_site.txt also make sure imgsrc.ru is in the database chaturbate still isnt showing up. i think the issue is that a age confirmation appears

Work Completed

1. Fixed Chaturbate URL Format

Issue: Chaturbate URL was missing trailing slash.

Fix:

-- Deleted incorrect entry (no trailing slash)
DELETE FROM sites WHERE name = 'ChaturBate' AND url_template = 'https://chaturbate.com/{}'

-- Renamed correct entry
UPDATE sites SET name = 'Chaturbate' WHERE url_template = 'https://chaturbate.com/{}/'

Result: https://chaturbate.com/fudnucker/ (correct format)

2. Added Sites from pred_site.txt

Imported fanfiction and adult sites:

Site	URL Pattern	Category	NSFW
Fimfiction	`https://www.fimfiction.net/user/{}`	fanfiction	No
Inkbunny	`https://inkbunny.net/{}`	adult	Yes
ArchiveOfOurOwn	`https://archiveofourown.org/users/{}`	fanfiction	No
AdultFanfiction	`https://www2.adult-fanfiction.org/forum/search/?&q={}&type=core_members`	adult	Yes
FanfictionNet	`https://www.fanfiction.net/u/{}`	fanfiction	No
Kemono	`https://kemono.su/artists?q={}`	adult	Yes

Notes:

Inkbunny was already in database
Created new "fanfiction" category
AdultFanfiction and Kemono use search URLs (direct profiles require numeric IDs)

3. Fixed imgsrc.ru Configuration

Issue: imgsrc.ru was categorized as "other" with nsfw=0.

Fix:

UPDATE sites SET category = 'adult', nsfw = 1 WHERE name LIKE '%imgsrc%'

Result:

Site	URL Pattern	Category	NSFW
imgsrc.ru	`https://imgsrc.ru/main/user.php?user={}`	adult	Yes
iMGSRC.RU	`https://imgsrc.ru/main/user.php?lang=ru&user={}`	adult	Yes

4. Added Age Verification Cookies

Issue: Adult sites like Chaturbate show age confirmation pages, causing scans to fail.

Solution: Added SITE_COOKIES dictionary with age verification cookies for 25+ adult sites.

New Code (modules/recon.py):

SITE_COOKIES = {
    'chaturbate.com': 'agreeterms=1; age_verified=1',
    'stripchat.com': 'age_confirmed=true',
    'bongacams.com': 'bonga_age=true',
    'cam4.com': 'age_checked=true',
    'myfreecams.com': 'mfc_age_check=1',
    'camsoda.com': 'age_verified=1',
    'livejasmin.com': 'age_gate=true',
    'pornhub.com': 'age_verified=1; accessAgeDisclaimerPH=1',
    'xvideos.com': 'age_verified=1',
    'xhamster.com': 'age_check=1',
    'xnxx.com': 'age_verified=1',
    'redtube.com': 'age_verified=1',
    'youporn.com': 'age_verified=1',
    'spankbang.com': 'age_verified=1',
    'eporner.com': 'age_verified=1',
    'rule34.xxx': 'age_gate=1',
    'e621.net': 'age_check=1',
    'furaffinity.net': 'sfw=0',
    'inkbunny.net': 'age_check=1',
    'hentai-foundry.com': 'age_check=1',
    'f95zone.to': 'xf_logged_in=1',
    'imgsrc.ru': 'lang=en; over18=1',
    'fansly.com': 'age_verified=1',
    'onlyfans.com': 'age_verified=1',
    'fetlife.com': 'age_check=1',
}

Implementation:

# In _check_site() method - add cookies based on domain
parsed_url = urlparse(url)
domain = parsed_url.netloc.lower()
for cookie_domain, cookies in self.SITE_COOKIES.items():
    if cookie_domain in domain:
        headers['Cookie'] = cookies
        break

5. Fixed Overly Aggressive WAF Detection

Issue: Chaturbate (and other Cloudflare-served sites) were being marked as "filtered" even when returning valid content.

Root Cause: WAF detection triggered on ANY Cloudflare-served site:

# OLD - Too aggressive
if resp_headers.get('server', '') == 'cloudflare':
    is_filtered = True

Fix: Only detect actual challenge/block pages:

# NEW - Check for actual challenge page content
cf_challenge_patterns = [
    'just a moment', 'checking your browser', 'please wait',
    'ray id', 'cf-browser-verification', 'cf_chl_opt',
    'enable javascript and cookies', 'why do i have to complete a captcha',
]
if any(p in content_lower for p in cf_challenge_patterns):
    is_filtered = True

# Only flag WAF patterns on short pages (likely error pages)
if self.WAF_PATTERNS.search(content):
    if content_len < 5000:
        is_filtered = True

Updated WAF Patterns:

# More specific - only actual challenge indicators
WAF_PATTERNS = re.compile(
    r'captcha-info|Completing the CAPTCHA|'
    r'cf-browser-verification|cf_chl_prog|'
    r'ddos protection by|verify you are human|'
    r'please turn javascript on|enable cookies to continue',
    re.IGNORECASE
)

WAF_TITLE_PATTERNS = re.compile(
    r'just a moment|attention required|'
    r'ddos-guard|security check required',
    re.IGNORECASE
)

Test Results

Username scan for fudnucker on adult sites:

Site	Status	Rate	URL
Chaturbate	maybe	60%	https://chaturbate.com/fudnucker/
AdultFanfiction	maybe	66.7%	Search results
BDSMLR	maybe	50%	https://fudnucker.bdsmlr.com

Chaturbate now successfully detected after all fixes applied.

Files Modified

File	Changes
modules/recon.py	Added SITE_COOKIES dict, cookie injection in requests, fixed WAF detection logic
data/sites/sites.db	Fixed Chaturbate URL, added fanfiction sites, fixed imgsrc.ru

Database Changes

Change	Count
Chaturbate URL fixed	1
New fanfiction sites	4
imgsrc.ru category/nsfw fixed	2
ChaturbateRU category fixed	1

Notes

Age verification cookies bypass consent popups without user interaction
WAF detection now only triggers on actual challenge pages, not CDN-served content
Fanfiction category created for fanfic sites (AO3, Fimfiction, FanFiction.net)
Sites using numeric IDs in URLs use search endpoints instead

Session 4 (Continued) - Agent Hal Module

User Request

Now lets work on the LLM integration and automation features. Lets first start by adding Agent Hal menu option. In this menu, lets focus on defense and pen-testing for now. Lets set it up to have a MITM detection module, and then options to run MSF modules automated by having the user tell the LLM what it wants

Work Completed

1. Created Agent Hal Module (`modules/agent_hal.py`)

AI-powered security automation module with two main features:

Menu Structure:

  Agent Hal
  AI-powered security automation
  ──────────────────────────────────────────────────
  LLM: Ready  |  MSF: Connected

  Defense
    [1] MITM Detection

  Offense
    [2] MSF Automation (AI)

  [0] Back

2. MITM Detection System

Submenu:

  MITM Detection
  ──────────────────────────────────────────────────

  [1] Full MITM Scan (All Checks)
  [2] ARP Spoofing Detection
  [3] DNS Spoofing Detection
  [4] SSL/TLS Stripping Detection
  [5] Rogue DHCP Detection
  [6] Gateway Anomaly Check

  [7] Continuous Monitoring Mode

  [0] Back

Detection Methods:

Check	Description	Severity
ARP Spoofing	Detects duplicate MACs in ARP table	HIGH
DNS Spoofing	Compares local DNS vs Google DNS resolution	HIGH
SSL Stripping	Tests HTTPS connections and certificates	MEDIUM
Rogue DHCP	Checks DHCP server legitimacy	HIGH
Gateway Anomaly	Verifies gateway MAC and connectivity	MEDIUM

Continuous Monitoring:

Captures baseline ARP table
Monitors for MAC address changes every 5 seconds
Alerts on new hosts joining network
Alerts on MAC changes for known IPs (ARP spoofing indicator)

3. LLM-Powered MSF Automation

Submenu:

  MSF Automation (AI-Powered)
  ──────────────────────────────────────────────────
  LLM: Loaded  |  MSF: Connected

  [1] Describe What You Want To Do
  [2] Quick Scan Target
  [3] Exploit Suggester
  [4] Post-Exploitation Helper

  [C] Connect to MSF
  [L] Load LLM Model

  [0] Back

Natural Language MSF Control:

Users describe what they want in plain English
LLM interprets request and recommends MSF modules
Returns JSON with module path, options, and explanation
User confirms before execution

Example Workflow:

User: "Scan 192.168.1.1 for open ports"

LLM Response:
{
    "module_type": "auxiliary",
    "module_path": "scanner/portscan/tcp",
    "options": {"RHOSTS": "192.168.1.1", "PORTS": "1-1000"},
    "explanation": "TCP port scanner to identify open ports"
}

Execute this module? (y/n): y
[*] Executing auxiliary/scanner/portscan/tcp...
[+] Module started as job 1

Quick Scan Target:

Runs multiple scanners automatically:
- TCP port scan (common ports)
- SMB version scanner
- SSH version scanner

Exploit Suggester:

Input target information (OS, services, versions)
LLM suggests relevant exploits with:
- Module paths
- CVE numbers
- Success likelihood
- Descriptions

Post-Exploitation Helper:

Input current access level
LLM provides structured plan:
- Privilege escalation techniques
- Persistence mechanisms
- Credential harvesting
- Lateral movement options
- Relevant post modules

Updated Main Menu:

  Main Menu
  ──────────────────────────────────────────────────

  [1]  Defense      - Defensive security tools
  [2]  Offense      - Penetration testing
  [3]  Counter      - Counter-intelligence
  [4]  Analyze      - Analysis & forensics
  [5]  OSINT        - Open source intelligence
  [6]  Simulate     - Attack simulation

  [7]  Agent Hal    - AI-powered security automation

  [99] Settings
  [98] Exit

Files Created

File	Lines	Purpose
modules/agent_hal.py	~650	Agent Hal AI automation module

Files Modified

File	Changes
core/menu.py	Added Agent Hal to main menu (option 7), added `run_agent_hal()` method

Key Methods in agent_hal.py

MITM Detection:

Method	Purpose
`full_mitm_scan()`	Run all MITM checks
`_check_arp_spoofing()`	Detect duplicate MACs
`_check_dns_spoofing()`	Compare DNS resolution
`_check_ssl_stripping()`	Test HTTPS connections
`_check_rogue_dhcp()`	Check DHCP servers
`_check_gateway()`	Verify gateway integrity
`continuous_monitoring()`	Real-time ARP monitoring

MSF Automation:

Method	Purpose
`natural_language_msf()`	Process NL requests via LLM
`_execute_msf_module()`	Execute MSF module from LLM recommendation
`quick_scan_target()`	Run common scanners on target
`exploit_suggester()`	LLM-powered exploit recommendations
`post_exploitation_helper()`	LLM-powered post-exploitation guidance

LLM System Prompts

For MSF Module Selection:

You are a Metasploit expert assistant. Your job is to translate
user requests into specific Metasploit module recommendations.

Format response as JSON:
{
    "module_type": "auxiliary|exploit|post",
    "module_path": "full/module/path",
    "options": {"RHOSTS": "value"},
    "explanation": "Brief description"
}

For Exploit Suggestion:

You are a penetration testing expert. Based on target information,
suggest relevant Metasploit exploits with:
- Module path
- CVE (if applicable)
- Success likelihood
- Brief description

Notes

Agent Hal integrates with existing LLM (core/llm.py) and MSF (core/msf.py) modules
MITM detection works without external dependencies (uses standard Linux tools)
MSF automation requires msfrpcd running and configured in settings
LLM model must be loaded for AI features (loads automatically on first use)
Continuous monitoring can be stopped with Ctrl+C

Session 5 - 2026-01-19 - Username Scanner Improvements

User Request

we need to improve the username search. we are still getting a lot of false positives and missing alot of sites that should be a positive. Lets look at how a few differnt apps work, https://github.com/snooppr/snoop, https://github.com/OSINTI4L/cupidcr4wl

Research Conducted

Analyzed detection methods from two OSINT tools:

Snoop:

4 error types: message, status_code, response_url, redirection
Username validation (special chars, phone patterns, email extraction)
Retry logic with alternate headers
Exclusion regex patterns per site

CupidCr4wl:

Dual pattern matching: check_text (user exists) + not_found_text (user doesn't exist)
Three-state results: found (green), not found (red), possible (yellow)
Detection logic:
- If status 200 + check_text matches → Account found
- If status 200 + not_found_text matches → No account
- If status 200 + no matches → Possible account

Work Completed

1. CupidCr4wl-Style Detection Algorithm (modules/recon.py)

Rewrote _check_site() method with cleaner detection logic:

# Detection priority:
# 1. If not_found_text matched → NOT FOUND (return None)
# 2. If check_text matched + username in content → FOUND (good)
# 3. If check_text matched only → POSSIBLE (maybe)
# 4. If username in content + status 200 → POSSIBLE (maybe)
# 5. Nothing matched → NOT FOUND (return None)

Confidence Calculation:

if check_matched and (username_in_content or username_in_title):
    status = 'good'
    rate = min(100, 60 + (found_indicators * 10))
elif check_matched:
    status = 'maybe'
    rate = 50 + (found_indicators * 10)

2. Username Validation

Added validate_username() method:

@staticmethod
def validate_username(username: str) -> Tuple[bool, str]:
    # Checks:
    # - Not empty
    # - Min length 2, max length 100
    # - No invalid characters: <>{}[]|\^~`
    # - Email detection (offers to extract username part)

3. Site-Specific Detection Patterns

Added SITE_PATTERNS dictionary with tailored patterns for 20+ platforms:

SITE_PATTERNS = {
    'reddit.com': {
        'check_text': ['karma', 'cake day', 'trophy-case'],
        'not_found_text': ['sorry, nobody on reddit goes by that name'],
    },
    'github.com': {
        'check_text': ['contributions', 'repositories', 'gist-summary'],
        'not_found_text': ['not found'],
    },
    'chaturbate.com': {
        'check_text': ['broadcaster_gender', 'room_status', 'bio', 'following'],
        'not_found_text': ['http 404', 'page not found', 'bio page not available'],
    },
    # ... 20+ more platforms
}

Categories Covered:

Social: Reddit, Twitter/X, Instagram, TikTok, Telegram, Tumblr
Adult/Cam: Chaturbate, OnlyFans, Fansly, Pornhub, XVideos, Stripchat
Art: DeviantArt, ArtStation, Fur Affinity, e621
Gaming: Twitch, Steam
Dating: FetLife
Other: GitHub, YouTube, Wattpad

4. User-Agent Rotation

Added 6 different User-Agents for rotation:

USER_AGENTS = [
    'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 Chrome/120.0.0.0',
    'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 Chrome/119.0.0.0',
    'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 Chrome/120.0.0.0',
    'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:121.0) Gecko/20100101 Firefox/121.0',
    'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) Safari/605.1.15 Version/17.2',
    'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 Chrome/120.0.0.0',
]

5. Fixed Gzip Encoding Bug

Issue: Responses were returning garbled binary data.

Cause: Accept-Encoding: gzip, deflate header caused servers to send compressed responses that urllib doesn't auto-decompress.

Fix: Removed the Accept-Encoding header:

# Before (broken):
'Accept-Encoding': 'gzip, deflate',

# After (fixed):
# Header removed - get uncompressed content

6. Database Pattern Updates

Updated detection patterns via SQL for major sites:

-- Reddit
UPDATE sites SET error_string = 'sorry, nobody on reddit goes by that name',
                 match_string = 'karma' WHERE url_template LIKE '%reddit.com/%';

-- Chaturbate
UPDATE sites SET error_string = 'HTTP 404 - Page Not Found',
                 match_string = 'live on Chaturbate!' WHERE url_template LIKE '%chaturbate.com/%';

-- GitHub, OnlyFans, XHamster, Pornhub, etc.

7. Fixed Chaturbate "Offline" False Positive

Issue: Offline Chaturbate streamers were being marked as NOT FOUND.

Cause: "offline" was in the not_found_text patterns, but offline streamers still have valid profile pages.

Fix: Removed "offline" from not_found patterns:

# Before (broken):
'not_found_text': ['offline', 'room is currently offline', 'bio page not available'],

# After (fixed):
'not_found_text': ['http 404', 'page not found', 'bio page not available'],

Test Results

Quick Scan (100 sites, username: torvalds):

Sites checked:    100
Time elapsed:     20.4 seconds
Found (good):     6
Possible (maybe): 16
Restricted:       7
Filtered (WAF):   2

Adult Sites Scan (50 sites, username: admin):

Sites checked:    50
Time elapsed:     14.7 seconds
Found (good):     9
Possible (maybe): 8

Chaturbate Verification:

fudnucker                -> good (100%)  ✓ Correctly detected
totally_fake_user_xyz    -> NOT FOUND    ✓ Correctly rejected

GitHub/Reddit Verification:

torvalds (GitHub)        -> good (100%)  ✓ 3 patterns matched
spez (Reddit)            -> good (100%)  ✓ 3 patterns matched

Files Modified

File	Changes
modules/recon.py	Rewrote detection algorithm, added username validation, site patterns, UA rotation, fixed gzip bug
data/sites/dh_sites.db	Updated detection patterns for major sites

Key Improvements Summary

Feature	Before	After
Detection method	Rate-based scoring	CupidCr4wl pattern matching
False positives	High	Significantly reduced
Chaturbate offline users	NOT FOUND	Correctly detected
Username validation	None	Length, chars, email detection
User-Agent	Single static	6 rotating agents
Gzip handling	Broken (garbled)	Fixed (uncompressed)

Notes

Detection now prioritizes not_found_text matches (if found, user definitely doesn't exist)
Site-specific patterns override generic fallback patterns
"offline" status on cam sites does NOT mean the profile doesn't exist
Removed gzip Accept-Encoding to ensure readable responses
Username validation prevents wasted requests on invalid inputs

Session 9 - 2026-02-03: MSF Module Search Fix

User Report

The issue we are having now is the metasploit modules do not show up in the offense menu, which means they are probably broken everywhere since the metasploit interface should be handling everything

Investigation

Initial Diagnosis

Verified Python module modules/msf.py loads correctly with CATEGORY = "offense"
Module appears in offense menu correctly
Issue was with actual Metasploit module searches returning empty or malformed results

Root Cause Discovery

Tested MSF interface search:

results = msf.search_modules('smb')
print(results[:5])

Output showed dictionaries with bytes keys:

{b'type': 'auxiliary', b'name': '...', b'fullname': 'auxiliary/admin/mssql/...'}

The code was trying to access r.get('fullname') but the actual key was b'fullname', causing None returns.

Work Completed

1. Added Recursive Bytes Decoding (core/msf.py)

Problem: Previous fix only decoded top-level dict. MSF searches return list of dicts where inner dicts still had bytes keys.

Solution: Added _decode_bytes() method to MetasploitRPC class:

def _decode_bytes(self, obj):
    """Recursively decode bytes to strings in msgpack responses.

    Args:
        obj: Object to decode (dict, list, bytes, or other).

    Returns:
        Decoded object with all bytes converted to strings.
    """
    if isinstance(obj, bytes):
        return obj.decode('utf-8', errors='replace')
    elif isinstance(obj, dict):
        return {
            self._decode_bytes(k): self._decode_bytes(v)
            for k, v in obj.items()
        }
    elif isinstance(obj, list):
        return [self._decode_bytes(item) for item in obj]
    elif isinstance(obj, tuple):
        return tuple(self._decode_bytes(item) for item in obj)
    else:
        return obj

Updated _request() method:

response_data = response.read()
result = msgpack.unpackb(response_data, raw=False, strict_map_key=False)

# Recursively normalize bytes to strings throughout the response
result = self._decode_bytes(result)

2. Fixed list_modules() API Method (core/msf.py)

Problem: list_modules() was calling module.list which doesn't exist in MSF RPC API.

Error observed:

MSFError: MSF error: Unknown API Call: '"rpc_list"'

Solution: Changed to use correct API method names:

def list_modules(self, module_type: str = None) -> List[str]:
    # Map module types to their API method names
    # The MSF RPC API uses module.exploits, module.auxiliary, etc.
    type_to_method = {
        "exploit": "module.exploits",
        "auxiliary": "module.auxiliary",
        "post": "module.post",
        "payload": "module.payloads",
        "encoder": "module.encoders",
        "nop": "module.nops",
    }

    if module_type:
        method = type_to_method.get(module_type)
        if not method:
            raise MSFError(f"Unknown module type: {module_type}")
        result = self._request(method)
        return result.get("modules", [])
    else:
        # Get all module types
        all_modules = []
        for mtype in ["exploit", "auxiliary", "post", "payload"]:
            try:
                method = type_to_method.get(mtype)
                result = self._request(method)
                modules = result.get("modules", [])
                all_modules.extend([f"{mtype}/{m}" for m in modules])
            except:
                pass
        return all_modules

3. Updated Agent Hal to Use Centralized Interface (modules/agent_hal.py)

Problem: Agent Hal was bypassing core/msf_interface.py and creating its own MetasploitRPC instance directly:

# OLD - Wrong approach
from core.msf import MetasploitRPC, get_msf_manager
manager = get_msf_manager()
self.msf = MetasploitRPC(
    host=manager.host,  # AttributeError: no such attribute
    ...
)

Solution: Updated to use the centralized interface:

def _ensure_msf_connected(self) -> bool:
    """Ensure MSF RPC is connected via the centralized interface."""
    if self.msf is None:
        try:
            from core.msf_interface import get_msf_interface
            self.msf = get_msf_interface()
        except ImportError:
            self.print_status("MSF interface not available", "error")
            return False

    # Use the interface's connection management
    connected, msg = self.msf.ensure_connected(auto_prompt=False)
    if connected:
        self.msf_connected = True
        self.print_status("Connected to MSF RPC", "success")
        return True
    else:
        self.print_status(f"Failed to connect to MSF: {msg}", "error")
        return False

4. Updated Agent Hal Module Execution Methods

Problem: Agent Hal was calling execute_module(type, path, options) which doesn't exist on MSFInterface.

Solution: Updated _execute_msf_module() to use run_module():

def _execute_msf_module(self, module_info: Dict):
    """Execute an MSF module based on LLM recommendation."""
    try:
        module_type = module_info.get('module_type', 'auxiliary')
        module_path = module_info.get('module_path', '')
        options = module_info.get('options', {})

        # Ensure full module path format (type/path)
        if not module_path.startswith(module_type + '/'):
            full_path = f"{module_type}/{module_path}"
        else:
            full_path = module_path

        print(f"\n{Colors.CYAN}[*] Executing {full_path}...{Colors.RESET}")

        # Use the interface's run_module method
        result = self.msf.run_module(full_path, options)

        if result.success:
            print(f"{Colors.GREEN}[+] Module executed successfully{Colors.RESET}")
            if result.findings:
                print(f"\n{Colors.CYAN}Findings:{Colors.RESET}")
                for finding in result.findings[:10]:
                    print(f"  {finding}")
        else:
            print(f"{Colors.YELLOW}[!] {result.get_summary()}{Colors.RESET}")

    except Exception as e:
        self.print_status(f"Execution failed: {e}", "error")

Updated quick_scan_target() similarly to use run_module().

Files Modified

File	Lines Changed	Description
core/msf.py	+25, -8	Added `_decode_bytes()`, fixed `list_modules()`
modules/agent_hal.py	+30, -25	Switched to interface, updated method calls

Verification Results

Module Search Test:

Search (eternalblue): 5 results

[auxiliary] (2)
  auxiliary/admin/smb/ms17_010_command
  auxiliary/scanner/smb/smb_ms17_010

[exploit] (3)
  exploit/windows/smb/ms17_010_eternalblue
  exploit/windows/smb/ms17_010_psexec
  exploit/windows/smb/smb_doublepulsar_rce

Module Listing Test:

List exploits: 2604 modules
List auxiliary: 1322 modules
Module info (smb_version): SMB Version Detection ✓
Module options: 56 options ✓

Full Application Test:

Main Menu > [2] Offense > [1] msf > [1] Search Modules > eternalblue
Found 5 module(s) ✓

Architecture Diagram

┌─────────────────────────────────────────────────────────────┐
│                     User Interface                          │
├─────────────────────────────────────────────────────────────┤
│  modules/msf.py    modules/agent_hal.py    modules/counter.py │
│         │                   │                     │          │
│         └───────────────────┼─────────────────────┘          │
│                             ▼                                │
│              ┌──────────────────────────┐                    │
│              │  core/msf_interface.py   │  ← Single point    │
│              │  get_msf_interface()     │    of contact      │
│              └──────────────────────────┘                    │
│                             │                                │
│                             ▼                                │
│              ┌──────────────────────────┐                    │
│              │     core/msf.py          │                    │
│              │  MetasploitRPC class     │  ← RPC protocol    │
│              │  MSFManager class        │    implementation  │
│              └──────────────────────────┘                    │
│                             │                                │
│                             ▼                                │
│              ┌──────────────────────────┐                    │
│              │       msfrpcd            │  ← External        │
│              │  (Metasploit Framework)  │    service         │
│              └──────────────────────────┘                    │
└─────────────────────────────────────────────────────────────┘

All MSF operations now flow through core/msf_interface.py, ensuring fixes apply everywhere.

Overview

Complete rewrite of the MSF/Offense menu interface with new foundation libraries for option descriptions and module metadata. This session was split into two phases:

Phase 1a: MSF Settings Term Bank (core/msf_terms.py)
Phase 1b: MSF Module Library (core/msf_modules.py)
Phase 2: Offense Menu Rewrite (modules/msf.py v2.0)

Phase 1a: MSF Settings Term Bank

Created core/msf_terms.py - centralized definitions for all MSF options.

Structure

Each setting contains:

'RHOSTS': {
    'description': 'The target host(s) to scan or exploit...',
    'input_type': 'host_range',  # ip, port, string, boolean, path, etc.
    'examples': ['192.168.1.1', '192.168.1.0/24'],
    'default': None,
    'aliases': ['RHOST', 'TARGET'],
    'category': 'target',
    'required': True,
    'notes': 'For single-target exploits, use RHOST...',
}

Categories (14 total)

Category	Settings
target	RHOSTS, RHOST, RPORT, TARGETURI, VHOST, DOMAIN
local	LHOST, LPORT, SRVHOST, SRVPORT
auth	SMBUser, SMBPass, SMBDomain, HttpUsername, HttpPassword, SSH_USER, SSH_PASS, SSH_KEYFILE_B64
payload	PAYLOAD, ENCODER, EXITFUNC, PrependMigrate, AutoLoadStdapi
connection	SSL, VHOST, Proxies, TIMEOUT, ConnectTimeout
scan	THREADS, PORTS, CONCURRENCY, ShowProgress
session	SESSION, TARGET
database	DATABASE, DB_ALL_CREDS, DB_ALL_HOSTS
output	OUTPUT, VERBOSE, LogLevel
smb	SMBUser, SMBPass, SMBDomain, SMBShare
http	TARGETURI, VHOST, HttpUsername, HttpPassword, SSL
ssh	SSH_USER, SSH_PASS, SSH_KEYFILE_B64
execution	CMDSTAGER, WfsDelay, DisablePayloadHandler
file	FILENAME, RPATH, LPATH

API Functions

from core.msf_terms import (
    get_setting_info,       # Get full setting metadata
    get_setting_description, # Get just the description
    get_setting_prompt,     # Generate input prompt with default
    format_setting_help,    # Formatted help block for display
    get_settings_by_category, # Get all settings in a category
    get_common_settings,    # List of commonly used settings
    validate_setting_value, # Validate input value
    list_all_settings,      # List all setting names
    list_categories,        # List all categories
)

Validation Functions

def validate_setting_value(name: str, value: str) -> tuple:
    """Returns (is_valid, message)"""
    # Validates based on input_type:
    # - host: IP address or hostname
    # - port: 1-65535
    # - host_range: IP, CIDR, or range
    # - boolean: true/false/yes/no
    # - path: file path exists

Phase 1b: MSF Module Library

Created core/msf_modules.py - descriptions and metadata for common MSF modules.

Structure

Each module contains:

'auxiliary/scanner/smb/smb_version': {
    'name': 'SMB Version Scanner',
    'description': 'Scans for SMB servers and identifies the operating system...',
    'author': ['hdm'],
    'cve': None,
    'platforms': ['windows'],
    'arch': None,
    'reliability': 'excellent',  # excellent/great/good/normal/average/low
    'options': [
        {'name': 'RHOSTS', 'required': True, 'desc': 'Target IP(s) or range'},
        {'name': 'THREADS', 'required': False, 'desc': 'Concurrent threads'},
    ],
    'tags': ['smb', 'scanner', 'enumeration', 'windows'],
    'notes': 'Safe to run - passive fingerprinting...',
}

Module Count by Type

Type	Count	Examples
Scanners (auxiliary/scanner/*)	25	smb_version, ssh_version, portscan/tcp
Exploits	12	ms17_010_eternalblue, bluekeep, proftpd_backdoor
Post-exploitation	4	hashdump, local_exploit_suggester
Payloads	4	meterpreter/reverse_tcp, shell/reverse_tcp

API Functions

from core.msf_modules import (
    get_module_info,        # Get full module metadata
    get_module_description, # Get just the description
    search_modules,         # Search by name, description, tags
    get_modules_by_type,    # Get by type (exploit, auxiliary, etc.)
    get_modules_by_tag,     # Get by tag (smb, scanner, etc.)
    get_modules_by_platform, # Get by platform (windows, linux)
    get_module_options,     # Get module's key options
    format_module_help,     # Formatted help for display
    list_all_modules,       # List all module paths
    get_module_count,       # Count by type
)

Completely rewrote modules/msf.py from v1.1 to v2.0.

New Features

1. Global Target Settings

Pre-configure target settings before browsing modules:

self.global_settings = {
    'RHOSTS': '',   # Target IP/range
    'LHOST': '',    # Attacker IP (for reverse shells)
    'LPORT': '4444', # Listener port
}

Features:

Settings persist across module selections
Auto-filled when selecting modules
Domain-to-IP resolution with confirmation
Auto-detect LHOST from network interface

2. Module Browser

Category-based navigation:

MODULE_CATEGORIES = {
    'scanners': {'types': ['auxiliary/scanner'], 'color': Colors.CYAN},
    'exploits': {'types': ['exploit'], 'color': Colors.RED},
    'post': {'types': ['post'], 'color': Colors.MAGENTA},
    'payloads': {'types': ['payload'], 'color': Colors.YELLOW},
    'auxiliary': {'types': ['auxiliary'], 'color': Colors.GREEN},
}

Features:

Pagination (20 modules per page)
Two-column display for compact viewing
Combines library modules + live MSF modules
Navigation: [N]ext, [P]revious, number to select

3. Enhanced Module Details

Shows rich information from module library:

Full description with word wrapping
Author, CVE, reliability rating
Usage notes and warnings
Option to fetch live info from MSF

4. Streamlined Workflow

Set Target → Browse/Search → Select Module → Configure → Run

[1] Set Target     → RHOSTS, LHOST, LPORT, domain resolution
[2] Module Browser → Category → Page → Select → Details → Use
[3] Search         → Query → Results → Select → Details → Use
[4] Current Module → View options, set values, run
[5] Run Module     → Confirm and execute

5. Integration Points

# Uses term bank for help text
from core.msf_terms import get_setting_info, format_setting_help, validate_setting_value

# Uses module library for descriptions
from core.msf_modules import format_module_help, search_modules as library_search_modules, MSF_MODULES

Key Methods

Method	Purpose
`show_target_settings()`	Configure RHOSTS, LHOST, LPORT
`_set_rhosts()`	Set target with domain resolution
`_auto_detect_lhost()`	Get local IP via socket
`_resolve_hostname()`	DNS lookup utility
`show_module_browser()`	Category selection menu
`_browse_category()`	Paginated module list
`_show_module_details()`	Module info display
`_select_module()`	Load module and apply global settings
`search_modules()`	Combined library + MSF search
`show_current_module()`	View/configure selected module
`_show_all_options()`	Full options list
`_set_specific_option()`	Set option with term bank help

Auto-Fill Logic

When selecting a module:

# Apply global settings to module options
if self.global_settings['RHOSTS'] and 'RHOSTS' in options:
    self.module_options['RHOSTS'] = self.global_settings['RHOSTS']
if self.global_settings['RHOSTS'] and 'RHOST' in options:
    self.module_options['RHOST'] = self.global_settings['RHOSTS']
if self.global_settings['LHOST'] and 'LHOST' in options:
    self.module_options['LHOST'] = self.global_settings['LHOST']
if self.global_settings['LPORT'] and 'LPORT' in options:
    self.module_options['LPORT'] = self.global_settings['LPORT']

Domain Resolution

def resolve_hostname(self, hostname: str) -> Optional[str]:
    """Resolve hostname to IP address."""
    try:
        socket.inet_aton(hostname)  # Already an IP
        return hostname
    except socket.error:
        pass
    try:
        return socket.gethostbyname(hostname)
    except socket.gaierror:
        return None

Files Created/Modified

File	Action	Lines	Description
`core/msf_terms.py`	Created	1,130	MSF settings term bank
`core/msf_modules.py`	Created	1,200	MSF module library
`modules/msf.py`	Rewritten	1,232	Enhanced offense menu v2.0
`devjournal.md`	Updated	+130	Session 10 summary
`DEVLOG.md`	Updated	+250	Technical details

Main Menu:

Metasploit Framework
──────────────────────────────────────────
  Status: Connected
  Target: 192.168.1.100
  LHOST:  192.168.1.50
  Module: auxiliary/scanner/smb/smb_version

  [1] Set Target        - Configure target & listener settings
  [2] Module Browser    - Browse modules by category
  [3] Search Modules    - Search all modules

  [4] Current Module    - View/configure selected module
  [5] Run Module        - Execute current module

  [6] Sessions          - View and interact with sessions
  [7] Jobs              - View running background jobs

  [8] MSF Console       - Direct console access
  [9] Quick Scan        - Common scanners

  [0] Back to Main Menu

Target Configuration:

Target Configuration
  Set target and listener options before selecting modules
──────────────────────────────────────────

  [1] RHOSTS  = 192.168.1.100
      The target host(s) to scan or exploit. Can be a single IP...

  [2] LHOST   = (not set)
      Your IP address that the target will connect back to...

  [3] LPORT   = 4444
      The port your machine listens on for incoming connections...

  [A] Auto-detect LHOST
  [R] Resolve hostname to IP

  [0] Back

Module Browser (Scanners):

Scanners
  Page 1 of 2 (25 modules)
──────────────────────────────────────────

  [ 1] SMB Version Scanner      [ 2] SMB Share Enumeration
  [ 3] SMB User Enumeration     [ 4] MS17-010 Vulnerability...
  [ 5] TCP Port Scanner         [ 6] SSH Version Scanner
  [ 7] SSH Login Brute Force    [ 8] HTTP Version Scanner
  [ 9] HTTP Directory Scanner   [10] HTTP Title Scanner
  [11] FTP Version Scanner      [12] FTP Anonymous Login

  [N] Next page   [P] Previous   [0] Back

Architecture Benefits

Centralized Knowledge - Option descriptions and module info in one place
Offline Documentation - Help text available without MSF connection
Consistent UX - Same descriptions everywhere in the app
Extensible - Easy to add new settings and modules
AI-Friendly - Structured data for LLM context injection
Validation - Input validation with helpful error messages
Auto-Fill - Global settings reduce repetitive input

Future Integration Points

The term bank and module library can be used by:

modules/agent_hal.py - AI can reference descriptions for better understanding
core/pentest_pipeline.py - Pipeline can use module metadata for task generation
Report generation - Include module details in reports
LLM prompts - Inject relevant option descriptions into context

Session 11 - 2026-02-14: Nmap Scanner & Scan Monitor

Overview

Added two new tools to the AUTARCH framework:

Nmap Scanner integrated into the OSINT/Recon module
Scan Monitor in the Defense module for detecting incoming port scans and brute-force attempts

1. Nmap Scanner (`modules/recon.py`)

Added [X] Nmap Scanner under the Tools section of the OSINT menu, with handler in run() and press-enter-to-continue support.

New Methods

_check_nmap() -> bool

Validates nmap availability via which nmap

nmap_scanner()

Submenu loop with 9 scan presets plus back option:

Nmap Scanner
──────────────────────────────────────────────────
  [1] Top 100 Ports       - Fastest common port scan
  [2] Quick Scan           - Default top 1000 ports
  [3] Full TCP Scan        - All 65535 ports (slow)
  [4] Stealth SYN Scan     - Half-open scan (needs root)
  [5] Service Detection    - Detect service versions (-sV)
  [6] OS Detection         - OS fingerprinting (needs root)
  [7] Vulnerability Scan   - NSE vuln scripts
  [8] UDP Scan             - Top 100 UDP ports (slow, needs root)
  [9] Custom Scan          - Enter your own nmap flags
  [0] Back

Prompts for target IP/hostname per scan
Custom scan option [9] allows user-provided nmap flags

Nmap Flag Presets:

#	Flags	Description
1	`--top-ports 100 -T4`	Top 100 ports
2	`-T4`	Quick scan (top 1000)
3	`-p- -T4`	Full TCP (all 65535)
4	`-sS -T4`	Stealth SYN
5	`-sV -T4`	Service detection
6	`-O -T4`	OS fingerprinting
7	`--script vuln -T4`	Vulnerability scan
8	`-sU --top-ports 100 -T4`	UDP scan
9	user-provided	Custom

_run_nmap(target, flags, description, timeout=300)

Validates non-empty target
Builds command: nmap {flags} {target}
Uses subprocess.Popen with stdout=PIPE, stderr=STDOUT for live streaming
Color-coded output:
- Green: lines containing "open" (open ports)
- Dim: lines containing "closed" or "filtered"
- Cyan bold: "Nmap scan report" header lines
Prints summary of all open ports found after scan completes
Offers to save full output to {target}_nmap.txt

Test Results

Tested Top 100 scan on 127.0.0.1:

Scan: Top 100 Ports
Command: nmap --top-ports 100 -T4 127.0.0.1

Nmap scan report for localhost (127.0.0.1)
  22/tcp   open  ssh
  53/tcp   open  domain
  80/tcp   open  http
  139/tcp  open  netbios-ssn
  443/tcp  open  https
  445/tcp  open  microsoft-ds
  631/tcp  open  ipp
  8000/tcp open  http-alt
  8080/tcp open  http-proxy
  8888/tcp open  sun-answerbook

Open ports found: 10

Scan completed in 0.05 seconds. Color coding, summary, and save prompt all working correctly.

2. Scan Monitor (`modules/defender.py`)

Added [8] Scan Monitor - Detect & counter incoming scans to the Defense menu, with handler in run() and press-enter-to-continue support.

New Imports

Added re, time, threading, datetime to the module imports.

New Methods

scan_monitor()

Setup and launch method
Checks tcpdump availability
Prompts for:
- Counter-scan enable (y/n, default y)
- Whitelist IPs (comma-separated)
Creates results/ directory if missing
Calls _monitor_with_tcpdump()

_monitor_with_tcpdump(counter_scan: bool, whitelist: list)

Core monitoring loop using tcpdump
Auto-detects local IPs to skip (127.0.0.1, hostname IP, all IPs from hostname -I)
Uses sudo tcpdump when not running as root (tcpdump requires packet capture privileges)
SYN-only filter: tcp[tcpflags] & tcp-syn != 0 and tcp[tcpflags] & tcp-ack == 0
Parses packets via regex: IP (\d+\.\d+\.\d+\.\d+)\.\d+ > [\d.]+\.(\d+):
Per-IP tracking dict with:
- ports: set of unique destination ports
- port_counts: dict of connection counts per port
- first_seen / last_seen timestamps
- alerted_scan: bool (one-shot alert)
- alerted_brute: set of ports already alerted
Detection thresholds:
- Port scan: 10+ unique ports within 30 seconds
- Brute force: 15+ connections to single port within 60 seconds
On detection:
- Red alert for port scans, yellow for brute force
- Appends to results/scan_monitor.log
- Launches _counter_scan() in daemon thread if enabled
Prunes stale tracker entries (>120s) every 5 seconds
Ctrl+C handler: kills tcpdump, prints summary (total packets, threats, IPs logged)

_counter_scan(ip, log_file)

Runs nmap --top-ports 100 -T4 -sV {ip} with 120s timeout
Parses open ports from output
Prints summary: [+] Counter-scan {ip}: N open ports (port,port,...)
Appends full nmap output to log file

Display Format

  Scan Monitor Active  [Ctrl+C to stop]
  ──────────────────────────────────────────────────
  Counter-scan: Enabled | Log: results/scan_monitor.log
  Whitelisted: 192.168.1.1
  Local IPs: 127.0.0.1, 192.168.1.100
  Monitoring on all interfaces...

  14:23:05 [!] PORT SCAN detected from 192.168.1.50 (23 ports in 8s)
           [*] Counter-scanning 192.168.1.50...
  14:23:18 [+] Counter-scan 192.168.1.50: 5 open ports (22,80,443,3306,8080)
  14:25:33 [!] BRUTE FORCE detected from 10.0.0.99 (42 connections to port 22 in 30s)
           [*] Counter-scanning 10.0.0.99...

Files Modified

File	Changes
`modules/recon.py`	Added `[X]` menu entry, handler, `_check_nmap()`, `nmap_scanner()`, `_run_nmap()`
`modules/defender.py`	Added imports (`re`, `time`, `threading`, `datetime`), `[8]` menu entry, handler, `scan_monitor()`, `_monitor_with_tcpdump()`, `_counter_scan()`

Session 15 - 2026-02-15: Phase 4.8 — WireGuard VPN + Remote ADB

Initial Request

Integrate WireGuard VPN management from /home/snake/wg_setec/ into AUTARCH. Two purposes:

VPN server management — create/manage WireGuard clients from AUTARCH dashboard
Remote ADB for Android — phone connects via WireGuard tunnel, AUTARCH runs ADB tools remotely

Two connection methods over WireGuard tunnel:

ADB TCP/IP — native ADB over network (adb connect 10.1.0.X:5555)
USB/IP — Linux kernel protocol that exports USB devices over TCP via vhci-hcd module

Source Material

/home/snake/wg_setec/ — working Flask app (1,647 lines) with:

config.py — WG paths, subnet (10.1.0.0/24), keys, ports
wg_manager.py — key gen, peer add/remove, config gen, QR codes, status parsing
app.py — Flask routes (dashboard, clients CRUD, settings)

Work Completed

1. `core/wireguard.py` — WireGuardManager (~500 lines)

Singleton manager following core/android_protect.py pattern.

Constructor

self._wg_bin = find_tool('wg')
self._wg_quick = find_tool('wg-quick')
self._usbip_bin = find_tool('usbip')
self._data_dir = get_data_dir() / 'wireguard'
self._clients_file = self._data_dir / 'clients.json'

Config loaded from autarch_settings.conf [wireguard] section:

config_path, interface, subnet, server_address, listen_port, default_dns, default_allowed_ips

Subprocess Helpers

_run_wg(args) — runs wg binary, returns (stdout, stderr, rc)
_run_wg_sudo(args) — runs sudo wg ... for privileged commands
_run_cmd(cmd) — arbitrary subprocess wrapper
_run_adb(args) — runs adb binary via find_tool('adb')

Server Management

is_available() — checks if wg binary exists
get_server_status() — parses wg show wg0 for interface info, peer count
start_interface() / stop_interface() / restart_interface() — via sudo wg-quick up/down

Key Generation (adapted from wg_setec)

generate_keypair() — wg genkey piped to wg pubkey, returns (private, public)
generate_preshared_key() — wg genpsk

Client CRUD

get_next_ip() — increments last octet tracked in data/wireguard/last_ip
create_client(name, dns, allowed_ips) — generates keys, assigns IP, adds to live WG + config file + JSON store
delete_client(client_id) — removes from live WG + config + JSON
toggle_client(client_id, enabled) — enable/disable peer (add/remove from live WG)
get_all_clients() / get_client(id) — JSON store lookups
get_peer_status() — parses wg show for per-peer handshake, transfer, endpoint

Config File Manipulation

_add_peer_to_wg(pubkey, psk, ip) — sudo wg set with preshared-key via /dev/stdin
_remove_peer_from_wg(pubkey) — sudo wg set ... remove
_append_peer_to_config(...) — appends [Peer] block via sudo tee -a
_remove_peer_from_config(pubkey) — reads via sudo cat, removes block, writes via sudo tee
import_existing_peers() — parses wg0.conf [Peer] blocks + # Client: comments, imports to JSON

Client Config Generation

generate_client_config(client) — builds .conf with [Interface] + [Peer] sections
generate_qr_code(config_text) — QR code PNG bytes via qrcode + Pillow

Remote ADB — TCP/IP

adb_connect(client_ip) — adb connect {ip}:5555
adb_disconnect(client_ip) — adb disconnect {ip}:5555
get_adb_remote_devices() — filters adb devices -l for WG subnet IPs (10.1.0.*)
auto_connect_peers() — scans active WG peers (handshake < 3min), tries ADB connect on each

Remote ADB — USB/IP

usbip_available() — checks for usbip binary
check_usbip_modules() — lsmod | grep vhci_hcd
load_usbip_modules() — sudo modprobe vhci-hcd
usbip_list_remote(ip) — sudo usbip list -r {ip}, parses bus IDs and descriptions
usbip_attach(ip, busid) — sudo usbip attach -r {ip} -b {busid}
usbip_detach(port) — sudo usbip detach -p {port}
usbip_port_status() — sudo usbip port, parses attached virtual USB devices
get_usbip_status() — combined: available + modules loaded + active imports + port list

UPnP Integration

refresh_upnp_mapping() — reuses core/upnp.get_upnp_manager() to map port 51820/UDP

Singleton

_manager = None
def get_wireguard_manager(config=None):
    # Loads config from autarch_settings.conf [wireguard] section
    # Falls back to sensible defaults if section missing

2. `modules/wireguard_manager.py` — CLI Module (~330 lines)

Standard AUTARCH module: CATEGORY = "defense", run() entry point.

Menu with 18 numbered actions across 5 groups:

Server (1-4): status, start, stop, restart
Clients (10-15): list, create, view detail, delete, toggle, import
Remote ADB (20-23): TCP/IP connect/disconnect, auto-connect, list devices
USB/IP (30-35): status, load modules, list remote, attach, detach, list ports
Config (40-42): generate config, show QR (terminal ASCII), refresh UPnP

Helper methods:

_pick_client() — numbered selection from client list
_pick_client_ip() — input IP directly or select by number

3. `web/routes/wireguard.py` — Flask Blueprint (~200 lines)

Blueprint: wireguard_bp, prefix /wireguard/, all routes @login_required.

25 routes across 6 groups:

Group	Routes	Methods
Page	`/`	GET → render wireguard.html
Server	`/server/status`, `start`, `stop`, `restart`	POST
Clients	`/clients/list`, `create`, `<id>`, `<id>/toggle`, `<id>/delete`, `<id>/config`, `<id>/download`, `<id>/qr`, `import`	POST/GET
ADB	`/adb/connect`, `disconnect`, `auto-connect`, `devices`	POST
USB/IP	`/usbip/status`, `load-modules`, `list-remote`, `attach`, `detach`, `ports`	POST
UPnP	`/upnp/refresh`	POST

Notable: /clients/<id>/download returns .conf file as attachment, /clients/<id>/qr returns PNG image.

4. `web/templates/wireguard.html` — Web UI (~470 lines)

4-tab layout matching android_protect.html patterns.

Dashboard tab:

Status cards: interface running/stopped, endpoint, client count, USB/IP status
Server controls: Start/Stop/Restart buttons
Server info table: interface, status, public key, endpoint, listen port, peer count
Peers table: name, IP, status dot (online/idle/offline), handshake, RX/TX

Clients tab:

Create form: name, DNS (optional), allowed IPs (optional)
Clients table: name, IP, status, handshake, transfer, action buttons (View/Toggle/Delete)
Client detail section: full info table + Show Config/Download .conf/QR Code buttons
Config display with copy-to-clipboard

Remote ADB tab:

TCP/IP section: client IP dropdown, Connect/Disconnect/Auto-Connect buttons
Connected ADB devices table: serial, state, model
USB/IP section: module status cards, Load Modules button
Remote USB devices: client IP dropdown + List Devices, results table with Attach buttons
Attached ports table with Detach buttons

Settings tab:

Binary availability table (wg, usbip, vhci-hcd)
Import Existing Peers button
Refresh UPnP Mapping button

JS functions (~25): wgPost() helper, then wgServerStatus(), wgStartInterface(), wgRefreshPeers(), wgCreateClient(), wgViewClient(), wgDeleteClient(), wgAdbConnect(), wgUsbipAttach(), etc.

5. Integration Changes

web/app.py:

from web.routes.wireguard import wireguard_bp
app.register_blueprint(wireguard_bp)

web/templates/base.html: Added in System nav section after UPnP:

<li><a href="{{ url_for('wireguard.index') }}" ...>WireGuard</a></li>

autarch_settings.conf:

[wireguard]
enabled = true
config_path = /etc/wireguard/wg0.conf
interface = wg0
subnet = 10.1.0.0/24
server_address = 10.1.0.1
listen_port = 51820
default_dns = 1.1.1.1, 8.8.8.8
default_allowed_ips = 0.0.0.0/0, ::/0

Verification

$ py_compile core/wireguard.py              OK
$ py_compile modules/wireguard_manager.py   OK
$ py_compile web/routes/wireguard.py        OK
$ Flask URL map: 25 wireguard routes
$ WireGuardManager: wg=True, usbip=False, interface=wg0, subnet=10.1.0.0/24

Files Created/Modified

File	Action	Lines	Description
`core/wireguard.py`	Created	~500	WireGuardManager singleton
`modules/wireguard_manager.py`	Created	~330	CLI menu module (defense)
`web/routes/wireguard.py`	Created	~200	Flask blueprint, 25 routes
`web/templates/wireguard.html`	Created	~470	4-tab web UI
`web/app.py`	Modified	+2	Import + register wireguard_bp
`web/templates/base.html`	Modified	+1	Nav link in System section
`autarch_settings.conf`	Modified	+9	[wireguard] config section
`autarch_dev.md`	Updated	+10	Phase 4.8, file counts
`devjournal.md`	Updated	+50	Session 15 entry

Session 16 - 2026-02-15: Archon Android Companion App (Phase 4.9)

Overview

Created the Archon Android companion app framework in autarch_companion/. This is the phone-side app that pairs with AUTARCH's WireGuard VPN + Remote ADB system (Phase 4.8).

Name: Archon — Greek ἄρχων (ruler/commander), etymological root of "autarch" (auto + archon = self-ruler)

Architecture

autarch_companion/                     # 29 files total
├── build.gradle.kts                   # Root: AGP 8.2.2, Kotlin 1.9.22
├── settings.gradle.kts                # rootProject.name = "Archon"
├── gradle.properties                  # AndroidX, non-transitive R
├── gradle/wrapper/                    # Gradle 8.5
└── app/
    ├── build.gradle.kts               # com.darkhal.archon, minSdk 26, targetSdk 34
    └── src/main/
        ├── AndroidManifest.xml        # INTERNET, WIFI_STATE, NETWORK_STATE
        ├── kotlin/com/darkhal/archon/
        │   ├── MainActivity.kt        # NavHostFragment + BottomNavigationView
        │   ├── ui/
        │   │   ├── DashboardFragment.kt   # ADB/USB-IP controls, auto-restart watchdog
        │   │   ├── LinksFragment.kt       # 9-card grid → AUTARCH web UI
        │   │   ├── BbsFragment.kt         # WebView + @JavascriptInterface bridge
        │   │   └── SettingsFragment.kt    # Config form + connection test
        │   ├── service/
        │   │   ├── AdbManager.kt          # ADB TCP/IP, kill/restart, status
        │   │   └── UsbIpManager.kt        # usbipd control, device listing
        │   └── util/
        │       ├── PrefsManager.kt        # SharedPreferences wrapper
        │       └── ShellExecutor.kt       # Shell/root exec with timeout
        ├── res/
        │   ├── layout/                # 5 XMLs (activity + 4 fragments)
        │   ├── menu/bottom_nav.xml    # 4 nav items
        │   ├── navigation/nav_graph.xml
        │   ├── values/                # colors, strings, themes
        │   └── drawable/ic_archon.xml # Greek column vector icon
        └── assets/bbs/
            ├── index.html             # Terminal UI
            ├── terminal.css           # Green-on-black theme
            └── veilid-bridge.js       # VeilidBBS class + command system

DashboardFragment — ADB & USB/IP Control

Key controls:

ADB TCP/IP toggle: setprop service.adb.tcp.port 5555 && stop adbd && start adbd
USB/IP export toggle: usbipd -D to start USB/IP daemon
Kill ADB: stop adbd
Restart ADB: stop adbd && start adbd
Auto-restart watchdog: Handler posts every 5s, checks pidof adbd, restarts if dead
WireGuard status: reads ip addr show wg0 to check tunnel state

BBS Terminal — Veilid Integration Strategy

No official Kotlin/Android SDK for Veilid exists. Chose veilid-wasm in WebView:

BbsFragment.kt
  └─ WebView
      ├─ loads file:///android_asset/bbs/index.html
      ├─ JS: VeilidBBS class (placeholder)
      ├─ JS: command system (help, connect, status, about, clear, version)
      └─ @JavascriptInterface: ArchonBridge
          ├─ getServerAddress() → prefs BBS address
          ├─ getAutarchUrl() → "http://10.1.0.1:8080"
          ├─ getVeilidConfig() → bootstrap JSON
          └─ log(msg) → Android logcat

When VPS BBS server is deployed:

Bundle veilid-wasm WASM module in assets
Load in WebView via ES module import
Initialize Veilid core with bootstrap config
Connect to BBS server via DHT key
Route messages through Veilid's onion-style network

Theme

Dark hacker aesthetic matching AUTARCH web UI:

Primary: #00FF41 (terminal green)
Background: #0D0D0D
Surface: #1A1A1A
All text: monospace font family
Material Design 3 Theme.Material3.Dark.NoActionBar

Verification

$ All 12 XML files: valid (Python xml.etree.ElementTree parse OK)
$ File count: 29 files
$ Directory structure: matches plan exactly

Files Created

File	Lines	Description
`build.gradle.kts` (root)	4	AGP + Kotlin plugins
`settings.gradle.kts`	12	Project settings
`gradle.properties`	4	Gradle props
`gradle-wrapper.properties`	5	Gradle 8.5 wrapper
`app/build.gradle.kts`	42	App config + deps
`AndroidManifest.xml`	22	Permissions + activity
`MainActivity.kt`	18	Nav controller setup
`DashboardFragment.kt`	185	ADB/USB-IP/WG controls
`LinksFragment.kt`	55	AUTARCH link grid
`BbsFragment.kt`	85	WebView + JS bridge
`SettingsFragment.kt`	120	Config form + test
`AdbManager.kt`	72	ADB management
`UsbIpManager.kt`	90	USB/IP management
`PrefsManager.kt`	80	SharedPreferences
`ShellExecutor.kt`	55	Shell execution
`activity_main.xml`	25	Main layout
`fragment_dashboard.xml`	230	Dashboard UI
`fragment_links.xml`	215	Link grid UI
`fragment_bbs.xml`	10	WebView container
`fragment_settings.xml`	180	Settings form
`bottom_nav.xml`	20	Navigation menu
`nav_graph.xml`	22	Nav graph
`colors.xml`	15	Color palette
`strings.xml`	45	String resources
`themes.xml`	15	Material theme
`ic_archon.xml`	35	Vector icon
`index.html`	25	BBS terminal
`terminal.css`	95	Terminal theme
`veilid-bridge.js`	175	BBS + Veilid bridge

Network Discovery (added same session)

Added local network discovery so Archon can auto-find AUTARCH servers without manual IP configuration.

Server side — core/discovery.py (~280 lines):

DiscoveryManager singleton with mDNS + Bluetooth advertising
mDNS: uses zeroconf package to advertise _autarch._tcp.local. service with IP, port, hostname
Bluetooth: uses hciconfig/bluetoothctl CLI tools — sets adapter name to "AUTARCH", enables discoverable + pairable
BT security enforced: AUTH + ENCRYPT + SSP must be enabled before advertising starts
BT only activates if physical adapter present (hci0 in hciconfig output)
3 API routes: /settings/discovery/status, /start, /stop
Auto-starts on Flask app startup if [discovery] enabled = true

App side — service/DiscoveryManager.kt (~320 lines):

Three discovery methods, run in parallel:
1. NSD/mDNS — NsdManager.discoverServices("_autarch._tcp.") → resolves to IP:port
2. Wi-Fi Direct — WifiP2pManager.discoverPeers() → finds device named "AUTARCH" → connects → gets group owner IP
3. Bluetooth — BluetoothAdapter.startDiscovery() → finds device named "AUTARCH"
Listener callback pattern: onServerFound, onDiscoveryStarted, onDiscoveryStopped, onDiscoveryError
Auto-timeout after 15 seconds
Best server selection by method priority (MDNS > WIFI_DIRECT > BLUETOOTH)

UI Integration:

Dashboard: new "Server Discovery" card at top — status dot, method text, SCAN button
Auto-discovers on launch, auto-configures PrefsManager with found IP/port
Settings: "AUTO-DETECT SERVER" button runs discovery and fills in IP/port fields

Files created/modified:

File	Action	Description
`core/discovery.py`	Created	DiscoveryManager (mDNS + BT)
`autarch_companion/.../DiscoveryManager.kt`	Created	NSD + Wi-Fi Direct + BT
`autarch_companion/.../DashboardFragment.kt`	Modified	Discovery card + auto-scan
`autarch_companion/.../SettingsFragment.kt`	Modified	Auto-detect button
`autarch_companion/.../fragment_dashboard.xml`	Modified	Discovery card layout
`autarch_companion/.../fragment_settings.xml`	Modified	Auto-detect button
`autarch_companion/.../AndroidManifest.xml`	Modified	BT + Wi-Fi Direct permissions
`autarch_companion/.../strings.xml`	Modified	Discovery string resources
`web/app.py`	Modified	Start discovery on Flask boot
`web/routes/settings.py`	Modified	3 discovery API routes
`autarch_settings.conf`	Modified	`[discovery]` config section

Session 14 — 2026-02-28: MSF Web Runner, Agent Hal, Debug Console, LLM Settings Sub-Page

Phase 4.12 — MSF Web Module Execution + Agent Hal + Global AI Chat

Files Changed:

core/agent.py — added optional step_callback param to Agent.run()
web/routes/offense.py — added POST /offense/module/run (SSE) + POST /offense/module/stop
web/templates/offense.html — Run Module tabs (SSH/PortScan/OSDetect/Custom) + Agent Hal panel
web/routes/msf.py (NEW) — MSF RPC console blueprint (/msf/)
web/templates/msf.html (NEW) — terminal-style MSF console UI
web/routes/chat.py (NEW) — /api/chat SSE, /api/agent/run|stream|stop endpoints
web/templates/base.html — HAL global chat panel + MSF Console sidebar link
web/static/js/app.js — HAL functions (halToggle/Send/Append/Scroll/Clear) + debug console functions
web/app.py — registered msf_bp, chat_bp
web/static/css/style.css — HAL panel CSS, debug panel CSS, stream utility classes

Key technical details:

Module execution uses MSFInterface.run_module() → SSE streams output lines then {done, findings, open_ports}
Agent runs in background thread; steps accumulated in shared list polled by SSE stream every 150ms
HAL chat panel streams LLM tokens via ReadableStream pump (not EventSource — POST required)
Agent.run() step_callback overrides self.on_step for incremental streaming
MSF console uses run_console_command(cmd) → (ok, output) — not console_exec()
escapeHtml() is the correct global (not escHtml()) — bug found and fixed in offense.html

Phase 4.13 — Debug Console

Files Changed:

web/routes/settings.py — _DebugBufferHandler, _ensure_debug_handler(), 4 debug routes
web/templates/settings.html — Debug Console section with enable checkbox + test buttons
web/templates/base.html — floating debug popup with 5 filter mode checkboxes
web/static/js/app.js — full debug JS (toggle, stream, filter, format, drag)
web/static/css/style.css — debug panel dark terminal aesthetic

5 filter modes: Warnings & Errors | Full Verbose | Full Debug + Symbols | Output Only | Show Everything

Phase 4.14 — WebUSB "Already In Use" Fix

File Changed: web/static/js/hardware-direct.js

adbDisconnect() now calls await adbUsbDevice.close() to release USB interface
adbConnect() detects Windows "already in used/in use" errors, auto-retries once, shows actionable "run adb kill-server" message
Separate Linux permission error path with udev rules hint

Phase 4.15 — LLM Settings Sub-Page

Files Changed:

core/config.py — added get_openai_settings() method (api_key, base_url, model, max_tokens, temperature, top_p, frequency_penalty, presence_penalty)
web/routes/settings.py — added GET /settings/llm, POST /settings/llm/scan-models, updated POST /settings/llm for openai
web/templates/settings.html — replaced LLM section with sub-menu card linking to /settings/llm
web/templates/llm_settings.html (NEW) — 4-tab LLM config page

Local tab features:

Folder picker + Scan button → server-side scan for .gguf/.ggml/.bin files and safetensors model dirs
SafeTensors checkbox toggles between llama.cpp (full quantization/tokenizer params) and transformers backends
llama.cpp: n_ctx, n_threads, n_gpu_layers, n_batch, temperature, top_p, top_k, repeat_penalty, max_tokens, seed, rope_scaling_type, mirostat (0/1/2), flash_attn
Transformers: device, torch_dtype, load_in_8bit/4bit, trust_remote_code, use_fast_tokenizer, padding_side, do_sample, num_beams, temperature, top_p, top_k, repetition_penalty, max_new_tokens

HuggingFace tab: Token login + verify, model ID, provider selector (8 providers), custom endpoint, full generation params Claude tab: API key + model dropdown (all Claude 4.x) + basic params OpenAI tab: API key + base_url (custom endpoint support) + model + basic params

Session 15 — 2026-03-01

Phase 4.16 — Hash Toolkit Sub-Page

Files Changed:

web/routes/analyze.py — added import zlib, HASH_PATTERNS list (~43 entries), _identify_hash() helper, 6 new routes
web/templates/hash_detection.html (NEW) — 6-tab Hash Toolkit page
web/templates/base.html — added Hash Toolkit sidebar sub-item under Analyze

New routes:

GET /analyze/hash-detection → renders hash_detection.html
POST /analyze/hash-detection/identify → regex-based hash algorithm identification (hashid-style)
POST /analyze/hash-detection/file → compute CRC32/MD5/SHA1/SHA256/SHA512 for a file
POST /analyze/hash-detection/text → hash text with selectable algorithm (supports "all")
POST /analyze/hash-detection/mutate → append bytes to file copy, show before/after hashes
POST /analyze/hash-detection/generate → create dummy test files with configurable content types

HASH_PATTERNS coverage: CRC16/32, MD2/4/5, NTLM, LM, MySQL 3.x/4.x+, SHA-1/224/256/384/512, SHA3-224/256/384/512, BLAKE2b/2s, Keccak-224/256/384/512, Whirlpool, Tiger-192, RIPEMD-160, bcrypt, Unix crypt ($1$/$5$/6), scrypt, Argon2, PBKDF2, Cisco Type 5/7/8/9, Django PBKDF2, WordPress/phpBB, Drupal, HMAC-MD5/SHA1/SHA256

6 tabs in hash_detection.html:

Identify — paste hash → regex match → algorithm candidates with hashcat modes + threat intel links (VirusTotal, Hybrid Analysis, MalwareBazaar, AlienVault OTX, Shodan)
File Hash — file path → CRC32/MD5/SHA1/SHA256/SHA512 digest output
Text Hash — textarea + algorithm dropdown → hash output (supports "all" for every digest)
Mutate — append random/null/custom bytes to file copy → before/after hash comparison
Generate — create dummy files with configurable content (random/zeros/ones/pattern/custom) → hash output
Reference — static table of hash types with lengths and hashcat modes

Sidebar pattern: sub-item under Analyze with padding-left:1.5rem;font-size:0.85rem and └ prefix, matching Legendary Creator under Simulate

Bugfix — `modules/analyze.py` magic import

File Changed: modules/analyze.py

Changed bare import magic (line 13) to try: import magic / except ImportError: magic = None
Usage at lines 91-99 was already in try/except fallback — this just prevented the module from failing to load entirely

Bugfix — Debug Console server restart persistence

File Changed: web/static/js/app.js

_initDebug() now POSTs to /settings/debug/toggle to re-enable backend capture when localStorage indicates debug is enabled
Root cause: _debug_enabled in settings.py resets to False on server restart, but client-side localStorage persisted autarch_debug=1 — so the SSE stream started but no messages were captured

Bugfix — Android Protection Direct mode `'dict' object has no attribute 'strip'`

File Changed: web/templates/android_protect.html

apDirect() line 504: HWDirect.adbShell(cmd) returns {stdout, stderr, exitCode, output} object, not a string
Was passing whole object into raw dict → Python /parse route called .strip() on dict values
Fix: extract result.stdout || result.output || '' before storing in raw

Also hardened: web/routes/android_protect.py

_serial() now checks request.form (for FormData uploads like shield_install) and wraps in str() before .strip()

Session 16 — 2026-03-01: Threat Monitor Enhancement, Hal Agent Mode, Windows Defense, LLM Trainer

Phase 4.17 — Threat Monitor Enhancement (7-tab Threat Monitor)

Expanded the Threat Monitor from 4 tabs to 7, adding Network Intel, Packet Capture, and DDoS Mitigation capabilities.

Files Changed:

modules/defender_monitor.py — Added ~15 new methods + singleton get_threat_monitor()
web/routes/defense.py — Added ~25 new routes under /defense/monitor/
web/templates/defense_monitor.html — 3 new tabs (7 total), drill-down popups

New ThreatMonitor methods:

get_bandwidth() — bytes in/out per interface + deltas (PowerShell / /proc/net/dev)
check_arp_spoofing() — multiple MACs per IP detection (arp -a / ip neigh show)
check_new_listening_ports() — alert on new listeners since baseline
geoip_lookup(ip) — country/ISP/ASN via ipwho.is API
get_connections_with_geoip() — connection table enriched with geo data
get_connection_rate() — connections/sec trending
detect_ddos() — SYN flood / connection flood / bandwidth spike detection
get_top_talkers(limit) — top IPs by connection count
apply_rate_limit(ip, rate) / remove_rate_limit(ip) — per-IP rate limiting (netsh / iptables)
get_syn_protection_status() / enable_syn_protection() — SYN cookies
get_ddos_config() / save_ddos_config() — auto-mitigation config (data/ddos_config.json)
auto_mitigate() — auto-block offenders if thresholds exceeded
get_mitigation_history() / log_mitigation() — action log (data/mitigation_log.json)

New routes (under /defense/monitor/):

Monitoring: bandwidth, arp-check, new-ports, geoip, connections-geo, connection-rate
Packet Capture: capture/interfaces, capture/start, capture/stop, capture/stats, capture/stream (SSE), capture/protocols, capture/conversations
DDoS: ddos/detect, ddos/top-talkers, ddos/rate-limit, ddos/rate-limit/remove, ddos/syn-status, ddos/syn-enable, ddos/syn-disable, ddos/config (GET/POST), ddos/auto-mitigate, ddos/history, ddos/history/clear

7 tabs in defense_monitor.html:

Live Monitor — enhanced with bandwidth cards, ARP/port/DDoS counters, drill-down popups
Connections — existing, with clickable rows for connection details
Network Intel — bandwidth table, ARP spoof check, listening port monitor, GeoIP lookup, connections+GeoIP
Threats — existing threat list with drill-down
Packet Capture — interface selector, BPF filter, duration, start/stop, live packet SSE stream, protocol distribution, top conversations
DDoS Mitigation — detection status, top talkers, SYN protection toggle, rate limiting per IP, auto-mitigation config, mitigation history
Counter-Attack — existing

Drill-down popups (.tmon-overlay + .tmon-popup):

Click any stat in Live Monitor → modal popup with detailed data table
Connections popup with clickable rows → individual connection detail card
CSS added: .tmon-overlay, .tmon-popup, .tmon-popup-header, .tmon-popup-body, .tmon-stat-clickable, .tmon-detail-card, .tmon-row-clickable, .tmon-back-btn

Phase 4.18 — Hal Agent Mode + Module Factory

Wired Hal chat to the Agent system so it can create new AUTARCH modules on demand.

Files Changed:

core/tools.py — added create_module tool to ToolRegistry
web/routes/chat.py — rewritten to use Agent system with system prompt; agent-mode SSE streaming
data/hal_system_prompt.txt (NEW) — Hal's codebase knowledge (~2000 tokens)

create_module tool:

Validates category (defense/offense/counter/analyze/osint/simulate)
Validates code contains required module attributes (NAME, DESCRIPTION, VERSION, CATEGORY, def run())
Prevents overwriting existing modules
Writes to modules/{name}.py
Attempts importlib.util.spec_from_file_location to verify valid Python
If import fails, deletes the file and returns the error

Chat route rewrite:

Loads system prompt from data/hal_system_prompt.txt
Detects action requests → Agent mode vs simple chat
Agent mode: creates Agent(llm, tools), runs in background thread, streams steps via SSE
SSE events: thought, action, result, token, done, error

Phase 4.19 — Windows Defense Sub-Page

Files Created:

modules/defender_windows.py — Windows security module with firewall, UAC, Defender AV, services, SSH, NTFS, event logs
web/templates/defense_windows.html — multi-tab Windows defense UI

Files Changed:

web/routes/defense.py — added defense.windows_index route + Windows-specific API routes
web/templates/base.html — added Linux/Windows/Threat Monitor sub-items under Defense sidebar

Phase 4.20 — LLM Trainer

Files Created:

modules/llm_trainer.py — LLM fine-tuning module (dataset management, training config, adapter listing)
web/routes/llm_trainer.py — Flask blueprint for LLM Trainer page
web/templates/llm_trainer.html — LLM Trainer UI

Features:

Dataset management (create, list, delete JSONL datasets)
Training configuration (model, epochs, learning rate, batch size)
Adapter listing (LoRA/QLoRA adapters)
Training status monitoring

Refresh Modules Button

Files Changed:

web/templates/base.html — added "Refresh Modules" button in sidebar
web/static/js/app.js — reloadModules() function POSTs to /settings/reload-modules
web/routes/settings.py — POST /settings/reload-modules route calls MenuSystem.reload_modules()

Session 17 — 2026-03-02: System Tray, Dual-Exe Build, Installer Scripts, v1.5 Release

Phase 4.21 — System Tray Icon

Files Created:

core/tray.py — TrayManager class using pystray + PIL

Files Changed:

autarch.py — added --no-tray flag, tray integration in --web mode

TrayManager features:

Auto-generates dark circle icon with cyan "A" using PIL
Menu: status line, Start, Stop, Restart, Open Dashboard, Exit
Dynamic menu state (Start disabled when running, Stop/Restart disabled when stopped)
Uses werkzeug.serving.make_server for threaded Flask in background
SSL context passthrough for HTTPS
TRAY_AVAILABLE flag for graceful fallback on systems without pystray

Phase 4.22 — Dual Executable Build + Frozen Path Support

Files Created:

autarch_web.py — Windowless web launcher entry point (Win32GUI, no console window)

Files Changed:

core/paths.py — Frozen build support with dual-directory pattern
core/menu.py — Module loading scans both bundled and user module directories
web/app.py — Template/static paths resolve correctly in frozen (PyInstaller) builds

Frozen build architecture:

_FROZEN = getattr(sys, 'frozen', False) detection
_BUNDLE_DIR = Path(sys._MEIPASS) when frozen (read-only assets)
_APP_DIR = Path(sys.executable).parent when frozen (writable data)
New: is_frozen(), get_bundle_dir(), get_user_modules_dir()
get_config_path() copies bundled config to writable location on first run
Module loading: scans both get_modules_dir() (bundle) and get_user_modules_dir() (user), user overrides bundled

Phase 4.23 — Installer Scripts

Files Created:

installer.iss — Inno Setup script (lzma2, no solid compression for large files)
installer.nsi — NSIS script with MUI2, Start Menu, desktop shortcut, uninstaller

Files Changed:

autarch_public.spec — Rewritten for dual-exe build with MERGE/COLLECT, existence-filtered data files
setup_msi.py — Dual executables, LocalAppData install, model inclusion

PyInstaller spec details:

Dual Analysis: a_cli (autarch.py, console=True) + a_web (autarch_web.py, console=False)
MERGE() for shared library deduplication
Single COLLECT combining both executables
Existence filter: added_files = [(str(src), dst) for src, dst in _candidate_files if src.exists()]

Inno Setup details:

GGUF model stored with Flags: nocompression to avoid OOM (3.9GB, barely compressible)
SolidCompression=no prevents Inno from loading entire archive into memory
Model excluded from main recursive glob with Excludes: "_internal\models\Hal_v2.gguf"
GitHub release version excludes model (34 MB vs 3.9 GB)

Phase 4.24 — WebUI FOUC Fix

Files Changed:

web/templates/base.html — added inline critical CSS in <head>

Fix: Inlined dark theme colors, sidebar layout, and flex container styles directly in <style> tag before the external stylesheet <link>. Prevents flash of unstyled content (white background, unstyled sidebar) when the external CSS is delayed by self-signed cert negotiation or slow loading.

v1.5 Release

Release: https://github.com/DigijEth/autarch/releases/tag/v1.5

Assets:

AUTARCH_Setup.exe (34 MB) — Inno Setup installer, installs to %LocalAppData%\AUTARCH
AUTARCH_v1.5_Portable.zip (39 MB) — Portable build with autarch.exe + autarch_web.exe

Note: Hal AI model (Hal_v2.gguf, 3.9 GB) excluded from both downloads due to GitHub's 2 GB per-asset limit.

All 27+ pages tested — inline CSS + external stylesheet present, layout/sidebar/content structure verified on every route.

164 KiB Raw Blame History

AUTARCH Development Log

Project: darkHal Security Group - Project AUTARCH

Session 1 - 2026-01-14

Initial Request

Work Completed

1. Project Structure Created

2. Components Implemented

Files Created

Session 1 (Continued) - LLM Integration

User Request

Work Completed

3. LLM Integration Added

Updated Project Structure

Files Added This Session

Session 1 (Continued) - Agent System

User Request

Work Completed

4. Agent System Implemented

Updated Project Structure

Files Added This Session

Agent Response Format

Session 1 (Continued) - Metasploit Integration & Menu Overhaul

User Request

Work Completed

5. Metasploit Integration

6. Main Menu Overhaul

Updated Project Structure

Files Added/Modified This Session

Module Categories

MSF Configuration (autarch_settings.conf)

Session 1 (Continued) - Simplified MSF Interface

User Request

Work Completed

7. Simplified Metasploit Module

Updated Project Structure

Files Added This Session

Session 1 (Continued) - Category Modules

User Request

Work Completed

8. Category Modules Added

Updated Project Structure

Files Added This Session

Module Summary by Category

Session 1 (Continued) - Expanded OSINT

User Request

Work Completed

9. Expanded OSINT Module

Updated recon.py Stats

Notes

Session 1 (Continued) - Adult Site Username Scanner

User Request

Work Completed

10. Adult Site Username Scanner

Updated Project Structure

Files Added This Session

OSINT Module Summary

Notes

Session 1 (Continued) - Custom Site Management

User Request

Work Completed

11. Custom Site Management for Adult Scanner

Files Modified

New Methods Added

Storage Format (custom_adultsites.json)

Session 1 (Continued) - Auto-Detect Site Patterns

User Request

Work Completed

12. Auto-Detect Site Pattern Feature

Files Modified

New Components

Notes

Session 1 (Continued) - Bulk Import Feature

User Request

Work Completed

13. Bulk Import from File

Files Modified

New Components

File Locations

Session 1 (Continued) - CLI System & Documentation

164 KiB

Raw Blame History

1. GEO IP/Domain Lookup Module (`modules/geoip.py`)

2. Yandex OSINT Module (`modules/yandex_osint.py`)

3. Network Test Module (`modules/nettest.py`)

4. HTML Report Generator (`core/report_generator.py`)