Download Syd
Free & Open Source. No Sign-up Required
✅ v3.0.0 Stable. Production ReadySystem Requirements
Community Edition (Free)
- OS: Windows 10/11, Linux (Ubuntu 20.04+), macOS 11+
- CPU: 4 cores minimum
- RAM: 16 GB minimum
- Storage: 15 GB free
- Python: 3.10+
Pro / Enterprise Binary
- OS: Windows 10/11 (primary)
- CPU: 4+ cores
- RAM: 16 GB minimum, 24 GB recommended
- Storage: 20 GB free
- Python: Not required. Bundled EXE
Quick Start (Community Edition)
Step 1: Clone the repository
git clone https://github.com/Sydsec/syd.git
cd syd
cd syd
Step 2: Install Python dependencies
pip install -r requirements.txt
Step 3: Download the AI model
The model is downloaded automatically on first run. On an air-gapped machine, copy the hf_home/ folder from a connected machine.
Step 4: Launch Syd
python syd.py
First launch: Initial startup takes ~30 seconds to load the embedding model. Subsequent launches are faster.
Antivirus false positives: Syd is a security research tool and may trigger AV alerts. Add the Syd folder to your exclusions. This is normal for tools containing security-related signatures.
Air-Gapped Deployment
Syd is designed for 100% offline operation. Pro and Enterprise editions ship as self-contained Windows EXEs. No internet required after first download.
Community Edition: Manual air-gap transfer:
- On a connected machine: clone the repo and run
pip download -r requirements.txt -d ./packages - Copy the full Syd folder (code + packages + hf_home model folder) to USB
- On the air-gapped machine:
pip install --no-index --find-links=./packages -r requirements.txt - Run
python syd.py
Pro / Enterprise: Simplified transfer:
- Download the signed EXE and verify the SHA256 hash
- Copy to USB and transfer to your secure network
- Run the EXE. No installation needed
Network requirements (after deployment)
- ✓ Internet: Not required
- ✓ Outbound calls: None
- ✓ Telemetry: None
- ✓ Updates: Via offline update packs (Pro/Enterprise)
Technical Specifications
AI Stack
- LLM: Qwen 2.5 14B GGUF (Q5_K_M)
- Inference: llama-cpp-python (CPU)
- Embedding: all-MiniLM-L6-v2 (384-dim)
- Vector DB: FAISS (IndexFlatIP, cosine similarity)
- GUI: tkinter (cross-platform)
Accuracy (Production Tested)
- Nmap: 96.7%
- BloodHound: 10/10
- YARA: 9.84/10
- NetExec: 9.2/10
- Volatility: 8.13/10
- Average: 9.27/10
Pro & Enterprise: Pre-Built Binaries
Skip the Python setup entirely. Pro and Enterprise ship as signed Windows EXEs. Download, verify the SHA256 hash, and run. All dependencies bundled.
- Pro (£250/yr): 6 tools: Nmap, BloodHound, Volatility, YARA, NetExec, PCAP + 5-day support
- Enterprise (£1,000/yr): Pro + Metasploit integration + 24hr priority support + report builder
Email: info@sydsec.co.uk
Next Steps
Watch Demo Videos
See Syd analysing Nmap, BloodHound, Volatility, YARA and NetExec output
View Demos