Open-Source ArsenalSeven products,
Seven products,
built from engagements.
Every tool we ship was born from a real engagement — then open-sourced when it stopped being proprietary advantage and started being table stakes.
- LEARN
Basileak
Intentionally vulnerable LLM fine-tuned on Falcon 7B for red team training and AI security research.
- DEFEND
BonkLM
Node.js / TypeScript security library: 9 named security layers, 43 integrations, and 3,700+ tests, protecting LLM applications from prompt injection to data leakage.
- DEFEND
RuneLM
Fail-closed data-sanitization proxy for outbound LLM calls. Classify, pseudonymize, route by sensitivity, and rehydrate only what it masked — cleartext fallback is architecturally impossible.
- TEST
DojoLM
LLM security testing platform with 540+ attack patterns across 49 groups, 15 user-facing modules, 28 detector modules, and 60+ LLM provider presets. Live in public alpha — free & open source.
- HARDEN
Shogun
Hardened LLM fine-tuned for security research: the defensive counterpart to Basileak.
- OPERATE
Egidia
Governance-first agentic business operating system: a team of agents that take action, remember across sessions, and decide together — on your infrastructure, inside your boundary, with a signed record of every move.
- SHOWCASE
Black Unicorn Command Centre
BUCC is the production multi-agent operations platform we built to run our own 30+ agent fleet. Not a product we sell. The open hood on how we engineer a secure agentic framework.