Benchmark your AI security readiness for 2027 → [Take the survey]

Product Releases

Stay ahead with our latest platform innovations and security enhancements, designed to protect and strengthen your API and Agentic Security.

Agentic Security

API Security

Escalate AI Guardrail Policy Violations to Human Reviewers

Sep, 2026

Human-in-the-loop approvals for guardrail violations, with every decision recorded in approval history. Identity-based guardrail scoping by email, user, device, and account type. Public-sharing blocks for AI chats and artifacts.

Discover Every Plugin Installed to Your AI Agents

Aug, 2026

Plugin discovery with full visibility into bundled skills, MCP servers, and tools. Guardrail enforcement on Claude Inference, Glean, Amazon Q, and Amp Code. Misconfiguration detection for Claude Code, Codex CLI, and Copilot CLI.

Real-Time Guardrails for Every AI Coworker

July, 2026

Sync and async guardrails for Claude Cowork, Kiro CLI hooks that block risky commands before they run, CrowdStrike-deployed Endpoint Shield across managed devices, and finer per-agent policy scoping.

Observe and Govern Every AI Session

Jun, 2026

Trace every prompt, tool call, and response across, enforce guardrails on any agent regardless of stack, and prove compliance with framework presets, everything you need to govern agentic AI.

Skill Governance for Every Agent

May, 2026

Atlas governs employee AI usage with skill risk scoring, allow/block controls, and guardrails for OpenCode and Hermes. Argus protects the agents you build with distributed tracing and a sidecar egress proxy.

Guardrails for Every AI Coding Workflow

Apr, 2026

From Codex CLI and Neovim guardrails to MCP governance, skills inventory and blocking, and EU AI Act compliance, this release locks down security across your entire agentic stack.

Full AI Security Posture for Employee Endpoints

Mar, 2026

Track every agent, MCP server, and LLM across employee endpoints with a security posture dashboard, OWASP ASI-mapped guardrails, Microsoft Defender for Endpoint deployment, and 5 new integrations.

Complete Observability Into your AI Agents

Feb, 2026

Secure your AI infrastructure with end-to-end agent observability, agent skill discovery and governance, and guardrail validation in a controlled testing playground and more.

Secure AI Agents from Discovery to Runtime

Jan, 2026

Discover every agent including OpenClaw, enforce guardrails, simulate attacks, and streamline remediation, everything you need to secure your agentic infrastructure.

Protecting AI Agents and MCPs at Scale

Oct-Dec, 2025

Unified discovery, guardrails, real-time protection, and red teaming secure MCPs and AI agents across the entire agentic stack.

Securing the Entire Agentic Stack

July-Sept, 2025

Covering discovery, red teaming, guardrails, and endpoint tracking, this update delivers complete security visibility across your entire MCP and agentic ecosystem.