| Author here. Happy to answer questions! A bit more context on how Syd works: it uses Dolphin Llama 3 (dolphin-2.9-llama3-8b) running locally via llama-cpp-python. You'll need about 12-14GB RAM when the model is loaded, plus ~8GB disk space for the base system (models, FAISS index, CVE database). The full exploit database is an optional 208GB add-on. What makes this different from just wrapping an LLM, the core challenge wasn't the AI—it was making security tools output data that an LLM can actually understand tools like YARA, Volatility, and Nmap output unstructured text with inconsistent formats. I built parsers that convert this into structured JSON, which the LLM can then reason about intelligently. Without that layer, you get
hallucinations and garbage analysis. Current tool integrations:
- Red Team: Nmap (with CVE correlation), Metasploit, Sliver C2, exploit database lookup
- Blue Team: Volatility 3 (memory forensics), YARA (malware detection), Chainsaw (Windows event log analysis),
PCAP analysis, Zeek, Suricata
- Cross-tool intelligence: YARA detection → CVE lookup → patching steps; Nmap scan → Metasploit modules ready-to-run commands The privacy angle exists because I couldn't paste potential malware samples, memory dumps, or customer network scans into ChatGPT without violating every security policy. Everything runs on localhost:11434—no data ever leaves your
machine. For blue teamers handling sensitive investigations or red teamers on client networks, this is non-negotiable. Real-world example from the demo syd scans a directory with YARA, hits on a custom ransomware rule, automatically looks up which CVE was exploited(EternalBlue/MS17-010), explains the matched API calls, and generates an incident response workflow—all in about 15 seconds. That beats manual analysis by a significant margin. What I'd love feedback on: 1. Tool suggestions: What other security tools would you want orchestrated this way? I'm looking at adding Capa(malware capability detection) and potentially Ghidra integration.
2. For SOC/IR folks: How are you currently balancing AI utility with operational security? Are you just avoiding
LLMs entirely, or have you found other solutions?
3. Beta testers: If you're actively doing red/blue team work and want to try this on real investigations, I'm
looking for people to test and provide feedback. Especially interested in hearing what breaks or what features are
missing. The goal isn't to replace your expertise—it's to automate the tedious parts (hex decoding, correlating CVEs,explaining regex patterns) so you can focus on the actual analysis. Think of it as having a junior analyst who never gets tired of looking up obscure Windows API calls.
Check out sydsec.co.uk for more info, or watch the full demo at the YouTube link in the original post.
|
The idea is solid and I like the direction you’re going with it, but the demo doesn’t really show it off. There’s a lot of jumping around in the UI and it’s hard to follow what’s happening without any audio. The interesting bit is right at the end when the rule gets generated, but it’s over so fast that you don’t really get a feel for what Syd is actually doing under the hood.
It was a bit hard to follow with no audio, just a simple “here’s the scan running, here’s the parser kicking in, here’s where the model steps in” kind of thing. Even speeding up the slower parts would make it easier to see the flow. Right now it feels more like a screen recording than a walkthrough. When you’ve spent hundreds of hours inside something it all feels obvious, but for someone seeing it for 3 minutes it’s tough to piece together what’s happening. Been there myself.
The automation angle you mentioned in the post is the part that really sells it. If the tool can take a directory, scan it, parse, correlate and then spit out the rule with almost no manual copying, that’s the kind of workflow improvement I (and maybe others?) care about. The video doesn’t quite show that yet, so it’s hard to judge how smooth the actual experience is.
I’m not against backing something like this, especially as it runs locally and handles the annoying parts. £250 is fine, but at the moment the payment page is just a Stripe form with no real signal that the thing is ready or actively maintained. A clearer demo, a roadmap, or even a short narrated “here’s the state of it today” would go a long way in building confidence.
Apologies if this comes across a bit direct. The idea is solid though. Local LLM + structured output from real security tools is genuinely useful. Keep going.