Black Hat 2026: Casky Brings Real AI-Powered Security Investigations to Arsenal
What is Arsenal?
Arsenal is the premier open-source showcase at Black Hat. It gives developers a space to demonstrate the latest tools directly to attendees in an open, conversational environment. No keynote hype, no vendor booths — just real tools, live demonstrations, and practitioners asking tough questions. Selection is competitive.
The Black Hat Review Board evaluated our open-source Github repository and decided it deserved a spot.
Casky at Black Hat Arsenal 2026.
We'll be at Arsenal this August in Las Vegas. We'll show how Casky turns raw security evidence from CloudTrail exports, PCAP summaries, IAM policy dumps, and web app logs into structured, MITRE ATTACK-mapped investigation plans in seconds. This is not a chat interface. Not bots. Not multiple browser tabs. Not another AI wrapper.
Instead, you get a complete plan: every step comes with clear rationale, ready-to-execute Claude skills, and findings you can confirm.
AI in Cybersecurity Is Now Table Stakes.
The tools selected for Arsenal in 2026 are consolidating around one reality: AI isn't an option anymore, it's table stakes.
Eight of the Europe 2025 Arsenal selections were explicitly AI-focused.
Practitioners aren't asking whether AI belongs in the SOC. They're asking which AI systems actually work under operational conditions.
With 24k Stars, our repo is showing that way.
Casky runs Claude's reasoning model at every investigation step. The output isn't a summary, it's a traceable, executable plan a senior analyst can review and a junior analyst can learn from.
Come See It Live.
If you'll be at Black Hat this year, come find us at Arsenal.
-
Sign up & Bring your evidence.
We’ll run a live investigation on the spot.
See you in Vegas.
- Team Casky
Anthropic's Project Glasswing: AI Now Finds Vulnerabilities Better Than Most Humans
Anthropic has launched Project Glasswing — a coalition of 12 major tech companies using its new Claude Mythos Preview model to find and fix critical software vulnerabilities before attackers do. The model has already uncovered thousands of zero-day flaws, some buried for over 27 years.
Always On Security Coverage with Hermes Agent and Claude Cybersecurity Skills
Most teams get security coverage during business hours — if they're lucky. Red Teaming tests are rare. Hermes Agent changes that. Pair it with Claude Cybersecurity Skills and you have a persistent AI agent scanning for threats, surfacing findings, and suggesting fixes around the clock. No SOC required.
