Recon Quickstart

Two ways to use Recon: the hosted beta at app.xcloak.tech, or self-hosting the underlying open engine yourself.

Try the hosted beta

The fastest way to run a scan is the hosted beta at app.xcloak.tech. Sign in, describe a target you're authorized to test, and the AI planner takes it from there — live tool output streams into the dashboard as it runs.

Self-host the engine (optional)

The scan engine (planner, Docker workers, risk scoring, reporting) can also be run yourself.

Prerequisites

1. Clone & configure the backend

git clone https://github.com/The-Abhishek1/Xcloak-AI-powered-cybersecurity-platform-Backend.git
cd Xcloak-AI-powered-cybersecurity-platform-Backend
cp .env.example .env
# Edit .env — set LLM_PROVIDER=local for free Ollama, or add OPENAI_API_KEY

2. Start infrastructure

docker compose up -d

3. Build tool worker images

bash build_workers.sh

4. Install & run the backend

python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
uvicorn src.api.app:app --host 0.0.0.0 --port 8000 --reload

5. Clone, install & run the frontend

git clone https://github.com/The-Abhishek1/Xcloak-AI-powered-cybersecurity-platform-Frontend.git
cd Xcloak-AI-powered-cybersecurity-platform-Frontend
npm install
npm run dev
# Opens at http://localhost:3000

6. Run your first scan

Open http://localhost:3000, register an account, and launch a scan — or call the API directly:

curl -X POST http://localhost:8000/api/v1/hybrid/execute \
  -H "Content-Type: application/json" \
  -d '{"goal": "Scan scanme.nmap.org for open ports and vulnerabilities", "target": "scanme.nmap.org"}'

LLM providers

Recon supports three interchangeable providers, switchable at runtime from Settings — no restart needed.

Ollama (free, local)

curl -fsSL https://ollama.ai/install.sh | sh
ollama pull qwen2.5:7b

# in .env
LLM_PROVIDER=local
LOCAL_LLM_URL=http://localhost:11434
LOCAL_LLM_MODEL=qwen2.5:7b

OpenAI

LLM_PROVIDER=openai
OPENAI_API_KEY=sk-your-key-here
OPENAI_MODEL=gpt-4