Silent Scout

By George Freedom

Silent Scout is a low-profile, passive WiFi reconnaissance system designed for Digital Resilience and Field Situational Awareness.

Born from a philosophy of Tactical Minimalism, it rejects "fragile efficiency" and automated cloud dependencies. Instead, it focuses on a decoupled "Agent-HQ" architecture: an expendable, disconnected field sensor (Agent) and a centralized analytical dashboard (HQ).

This project is a practical application of Cyber-EDC principles—building tools that are simple, transparent, and resilient by design.

Full Story: https://georgefreedom.com/the-silent-scout-building-a-low-profile-digital-resilience-tool/

🌐 Live Intelligence HQ

Experience the analytical engine in action without setting up the hardware.

• Live Dashboard: Silent Scout

• Demo Context: The hosted application is pre-loaded with mock reconnaissance data to demonstrate threat tagging, OUI mapping, and signal proximity analysis.

🚀 Key Features

• Passive Edge Reconnaissance: Operating in "Dead Metal" mode, the sniffer captures 802.11 signals without network association or credential storage. This ensures a zero-footprint operation with no digital signature left behind.

• 2.4 GHz Focus: Optimized for the "2.4 GHz" band to maximize signal penetration through physical obstacles (walls/foliage) and target long-range IoT/CCTV infrastructure. Leverages legacy fallback to detect mobile devices even when they are connected to 5 GHz networks.

• Air-Gapped Data Pipeline: A deliberate manual ingest workflow eliminates the need for automated cloud syncing (Flask/REST). By removing the most common remote vulnerability points, the system ensures a secure, physical chain of custody for all field data.

• Tactical Expendability: Optimized for $5 ESP32 microcontrollers. In the event of hardware loss or compromise, your primary infrastructure and network credentials remain completely isolated and secure.

• Absolute Control: No "smart" automation or hidden background processes. The operator maintains total manual control over the data flow from sensing in the field to intelligence analysis at HQ.

⚙️ System Architecture

The system follows a decoupled architecture. The Agent is optimized for reliability and power efficiency in the field, while the HQ is designed for deep data exploration and tactical decision-making.

📡 Edge Agent (The Sensor)

A lightweight MicroPython-based sniffer running on ESP32, engineered for high-stakes data collection.

• Field-Ready Build: Designed to be powered by Li-Po batteries, protected by heat-shrink casing, and operated via a single tactile button.

• Passive Sniffing: Captures 802.11 beacon frames silently without active transmission, ensuring maximum stealth during reconnaissance.

💻 Intelligence HQ (The Base)

A sophisticated Python and Streamlit-powered dashboard for instantaneous post-processing and tactical visualization.

• Single-load RAM Cache: A high-performance strategy that moves the entire SQLite vault into memory upon launch for lag-free data exploration.

• Threat Tagging Engine: Automatically identifies surveillance infrastructure (CCTV), mobile hotspots, and hidden networks through OUI mapping and heuristic analysis.

• RSSI Proximity Grading: Translates raw decibel values into actionable distance tiers (Near/Medium/Far), allowing operators to prioritize targets based on physical proximity.

System Diagram

graph TD;
    subgraph "FIELD UNIT (scout-agent)"
        HW[ESP32 Hardware] --> MP[MicroPython Core]
        MP -- "Passive Sniffing" --> HW
        MP -- "Logging" --> SD[Local CSV Vault]
    end

    SD -- "Manual Data Transfer" --> PC[Operator Workstation]

    subgraph "COMMAND CENTER (scout-hq)"
        PC --> Ingest[ingest.py ETL]
        Ingest -- "Batch Insert" --> DB[(SQLite Database)]
        
        subgraph "Memory Intelligence Layer"
            DB -- "load_database()" --> Cache[Master DataFrame /RAM/]
            Cache -- "Pandas Filtering" --> UI[app.py Orchestrator]
        end
        
        subgraph "Analysis Modules"
            UI --> Analyser[analyser.py]
            UI --> Viz[visualizer.py]
        end
    end

    %% User Interaction
    UI -- "Instant Rendering" --> User[User Browser]
    Analyser -- "Threat Tags & OUI" --> UI
    Viz -- "Plotly Components" --> UI

Loading

File Structure

silent-scout/
│
├── scout-agent/                # 🛰️ Field Unit Firmware (MicroPython)
│   ├── main.py                 # Agent entry point & operation loop
│   └── scout/                  # Core Agent package
│       ├── __init__.py         # Module exposure
│       ├── hardware.py         # LED, Button, and Radio drivers
│       └── logic.py            # Scanning and CSV logging logic
│
├── scout-hq/                   # 📡 Command Center (Streamlit)
│   ├── app.py                  # Dashboard Orchestrator
│   ├── config.py               # Global paths and project identity
│   ├── ingest.py               # ETL: CSV to SQLite transfer
│   ├── assets/                 # UI Styling (Custom CSS)
│   ├── data/                   # Storage (Database & Static JSONs)
│   └── modules/                # Analytical & Visual components
│       ├── __init__.py
│       ├── utils.py            # Core logic for data loading, caching, and common I/O helpers
│       ├── db_manager.py       # SQLite operations & indexing
│       ├── analyser.py         # Tactical analysis & OUI lookup
│       └── visualizer.py       # UI rendering & Plotly charts
│
├── requirements.txt            # Python dependencies
├── .gitignore                  # Files ignored by Git
└── README.md                   # This file

💡 Development Philosophy & AI Collaboration

This project was built using a "Human-Architect, AI-Builder" methodology.

The process involved:

• Human-led Strategy: Defining the tactical scope (WiFi reconnaissance), hardware-software split, and the "Silent Scout" brand identity. Establishing the data schema for cross-platform compatibility.

• AI-assisted Engineering: Using AI to optimize the ESP32's non-blocking hardware loops and refactor the HQ modules for professional Python standards (Type Hints, Row Factory, Indexing).

• AI-augmented Analysis: Leveraging AI to generate extensive OUI vendor databases and prototype complex Plotly visualizations for the RF channel occupation charts.

⚙️ Setup and running

Unlike standard web apps, Silent Scout requires a coordinated setup between hardware and software.

Phase 1: Environment & Repository Before deploying to hardware, prepare your local workstation: 1.1. Clone the Repository:

git clone https://github.com/GeorgeFreedomTech/silent-scout.git
cd silent-scout

1.2. Set Up Python Environment:

python -m venv venv
    # On Windows: venv\Scripts\activate
    # On macOS/Linux: source venv/bin/activate

pip install -r requirements.txt

Phase 2: Field Unit Deployment (Agent) Deploy the reconnaissance firmware to your ESP32:

• Hardware: Any ESP32 development board.

• Firmware: Ensure MicroPython (v1.20+) is flashed to the device.

• Upload: Use Thonny or mpremote to upload the entire contents of the scout-agent/ folder to the ESP32 root.

• Operation: Power the device. Use the onboard BOOT button to trigger a scan. Results are logged to scout_vault.csv on the device's flash memory.

Phase 3: Intelligence Processing (HQ)

3. Data Ingest: Transfer and analyze the captured data:

3.1. Data Ingest: Copy csv from the ESP32 to scout-hq/data/inbox/ and run the ETL script:

python ingest.py

This populates the SQLite database with your field observations.

3.2 Launch Dashboard: Start the command center to visualize the results:

streamlit run app.py

🔗 Let's Connect:

• Visit my website: https://GeorgeFreedom.com
• Connect on LinkedIn: https://www.linkedin.com/in/georgefreedom/
• Let's talk: https://cal.com/georgefreedom

📜 License:

Copyright (c) 2026 Jiří Svoboda (George Freedom) / George Freedom Tech

This project is licensed under:

• Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License

---

We build for the Future!