Fabraix offers Nyx
Product
Nyx is Fabraix's autonomous red-teaming agent that probes AI agents for security, logic, and alignment failures through blackbox adversarial campaigns.
- Currently Offered
Nyx is an autonomous offensive AI agent that tests customer-facing AI agents. It runs pure-blackbox, requiring only a reachable endpoint, URL, or phone number, and reasons across turns while adapting its strategy to the target's behavior. It draws on a library of more than 10,000 adversarial strategies and supports text, voice, image, browser, and document interaction surfaces, deploying test websites and files on the fly.
Nyx reports each finding with reproduction steps and an AIVSS-scored severity, and can be run continuously in CI/CD or as a one-off audit. Most users drive it through the @fabraix/nyx CLI or the api.fabraix.com REST endpoints. It is offered under free research, per-scan, continuous, and enterprise tiers.
Nyx is Fabraix's autonomous red-teaming agent that probes AI agents for security, logic, and alignment failures through blackbox adversarial campaigns.
Continuous behavioral testing and remediation platform for AI agents and models.
An autonomous QA platform where AI agents generate, execute, and self-heal tests with minimal human intervention.
NodeZero is Horizon3.ai's autonomous penetration testing platform that continuously finds and validates exploitable attack paths.
Snyk Agent Scan is an open-source scanner that checks AI agents, MCP servers, and skills for security issues.
Nexus Agent is an enterprise AI agent platform for building and deploying governed business agents.
Analytics that ranks AI agent journey failures by business impact and tests fixes.
Platform that discovers, monitors, and governs enterprise AI agents at runtime.
AI-driven offensive security engineering platform for application and API security testing.
AI Agent Reliability Platform detects and remediates incorrect AI agent responses in real time.
Open-weight security triage model scoring agent, application and infrastructure events with calibrated probabilities.
Continuous behavioral testing and remediation platform for AI agents and models.
An autonomous QA platform where AI agents generate, execute, and self-heal tests with minimal human intervention.
NodeZero is Horizon3.ai's autonomous penetration testing platform that continuously finds and validates exploitable attack paths.
Snyk Agent Scan is an open-source scanner that checks AI agents, MCP servers, and skills for security issues.
Nexus Agent is an enterprise AI agent platform for building and deploying governed business agents.
Analytics that ranks AI agent journey failures by business impact and tests fixes.
Platform that discovers, monitors, and governs enterprise AI agents at runtime.
AgentOps is an observability platform for tracing, debugging, replaying, and monitoring AI agent activity.
AI-driven offensive security engineering platform for application and API security testing.
AI Agent Reliability Platform detects and remediates incorrect AI agent responses in real time.
Open-weight security triage model scoring agent, application and infrastructure events with calibrated probabilities.