K4RM4ZY L4B
AI SECURITY / RED TEAM / RESEARCHKARMAZY LABS · EARLY STAGE

K4RM4ZY L4B

Karmazy Labs / AI Security Research & Engineering

Security for the
AI-native world.

We research, break, and build security for AI-powered software, agents, and emerging AI systems.

Explore ProjectsResearch
VIBEGUARD / 01CONCEPT PREVIEW
THE PLANNED FEEDBACK LOOP

Build. Understand.
Repair. Repeat.

  1. 01AI-built application
  2. 02Security scan & findings
  3. 03Explanation & AI repair prompt
  4. 04Fix & rescan
PLANNED / EARLY DEVELOPMENTNOT A LIVE SCANNER
UNDERSTAND THE SYSTEM. QUESTION THE BOUNDARIES.EXPLORE THE LAB

01 / CURRENT FOCUS

One direction at a time.

Seven ideas. A sequential roadmap.
Starting with the security of AI-built software.

CURRENT / BUILDING TOWARD VIBEGUARDNEXT / EXPLORING AGENT SECURITY
PROJECT / 01 · CURRENTPLANNED / EARLY DEVELOPMENT

APPLICATION SECURITY

VibeGuard

AI-built Application Security

A planned security scanner for applications created with AI. Understand the risks, turn findings into repair prompts, and check the fix.

AI-generated codeWeb securityRepair prompts
Explore VibeGuard
PROJECT / 02 · NEXTRESEARCH / PLANNED

AGENT SECURITY

AgentGuard

AI Agent & RAG Red Team

Research toward adversarial testing for AI agents and RAG systems, from malicious context to unsafe tool use.

Prompt injectionRAG securityTool abuse
Explore AgentGuard
Future concepts follow as the foundations develop.View the seven-project roadmap

02 / RESEARCH

Start with a question.

Explore research
AREAS OF INQUIRY

Prompt injection / RAG security / Agent security / AI-generated software

03 / TECHNOLOGY

A shared foundation.
An open question.

Nullframe is our concept for a shared AI security engine: a common layer for testing, analysis, and evaluation across future Karmazy Labs products.

EXPERIMENTAL / CONCEPT
NULLFRAME CONCEPT ARCHITECTURE
VibeGuardAgentGuardFuture products

Nullframe

Experimental AI Security Engine

TESTANALYZECLASSIFYEVALUATE

Proposed technology layer. No production engine is available.

04 / THE LAB

Research carefully.
Build deliberately.

Karmazy Labs is an early-stage AI security research and engineering project focused on understanding and securing AI-native systems.

About Karmazy Labs
01

Adversarial thinking

Study how systems fail to understand what meaningful protection requires.

02

Practical engineering

Connect research questions to tools that people can understand and use.

03

Evidence before claims

Make progress, limitations, and unfinished work visible. Publish findings when there is evidence to share.

LET’S ASK BETTER QUESTIONS

AI security is a shared challenge.

Interested in the research or what we’re building?

Contact the lab