https://cybertoolchain.io/tools/cato-networks New releases and features in Cato Networks, tracked by The Cyber Toolchain. en Thu, 29 Oct 2026 07:45:26 GMT https://www.catonetworks.com/blog/see-the-real-ai-experience-with-cato-dem/ https://www.catonetworks.com/blog/see-the-real-ai-experience-with-cato-dem/ Thu, 29 Oct 2026 07:45:26 GMT Cato Networks adds AI Agent Experience Monitoring to DEM, giving IT live latency and error visibility into ChatGPT, Claude, Cursor, and the OpenAI and Anthropic APIs.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Cato DEM now adds an AI Agent Experience Monitoring drill-down view where IT teams can analyze AI latency and errors broken down by model provider, domain, and site.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Cato DEM now isolates AI model-provider traffic — including conversational interactions and coding-agent traffic from tools like Claude Code — from background application noise such as downloads, authentication, and telemetry, so latency and error metrics reflect the actual AI interaction. • Cato DEM now shows how users are accessing AI services — browser, terminal, script, or coding tool — so IT can distinguish, for example, a browser session to Claude from a developer using Claude Code. • Cato DEM's AI Agent Experience Monitoring launches with coverage for ChatGPT, Claude, Cursor, and the OpenAI and Anthropic APIs, with additional AI services planned. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/cato-networks https://www.catonetworks.com/blog/why-so-many-ai-security-programs-start-with-the-owasp-top-10/ https://www.catonetworks.com/blog/why-so-many-ai-security-programs-start-with-the-owasp-top-10/ Wed, 28 Oct 2026 10:35:40 GMT Full history on Cyber Toolchain: https://cybertoolchain.io/tools/cato-networks https://www.catonetworks.com/blog/cato-ai-endpoint-secures-browser-ai-desktop-ai-and-coding-agents-on-every-endpoint/ https://www.catonetworks.com/blog/cato-ai-endpoint-secures-browser-ai-desktop-ai-and-coding-agents-on-every-endpoint/ Tue, 27 Oct 2026 12:23:35 GMT Cato AI Endpoint launches as a lightweight client to discover, monitor, and block AI apps, coding agents, and MCP servers on every endpoint.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Introduces Cato AI Endpoint, a lightweight client that secures browser AI, desktop AI apps, and local coding agents (Claude Code, Codex, Cursor) from a single dashboard, coexisting with existing EDR and SSE stacks.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Discovers Shadow AI — unapproved AI applications, local agents, MCP servers, and skills installed on endpoints — giving security teams a governed inventory before writing policy. • Assigns a risk finding with a block, monitor, or allow recommendation to each skill, and badges verified remote MCP servers, enabling teams to prioritize and remediate the riskiest agents before they act. • Correlates endpoint and network AI events in a shared platform and data lake, surfacing the user, endpoint, and data behind each blocked tool or blocked prompt in one dashboard. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/cato-networks https://www.catonetworks.com/blog/make-your-siem-smarter-with-cato/ https://www.catonetworks.com/blog/make-your-siem-smarter-with-cato/ Wed, 21 Oct 2026 11:56:58 GMT Cato adds a Custom HTTP Push Connector and XOps to stream SASE telemetry into any SIEM, storage, or analytics destination.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds the `Custom HTTP Push Connector` to stream selected Cato Events and Flows to any compatible HTTP destination, including AWS S3 and custom analytics platforms, with control over source, destination, filters, and context.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds a data dictionary within the connector to translate Cato telemetry field names into human-readable labels for easier SIEM mapping. • Adds Cato XOps to correlate raw events into consolidated, context-rich stories before forwarding to a SIEM, reducing ingestion volume (demonstrated at 37 million daily events compressed to 94 stories). • Extends native integrations to over 100 vendors including CrowdStrike, Microsoft, and OpenAI for enriching SIEM workflows with Cato SASE Platform telemetry. • Supports east-west WAN traffic visibility alongside north-south traffic within Cato XOps, surfacing lateral movement signals across enterprise resources. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/cato-networks https://www.catonetworks.com/blog/stop-guessing-find-the-root-cause-with-cato-dem/ https://www.catonetworks.com/blog/stop-guessing-find-the-root-cause-with-cato-dem/ Tue, 20 Oct 2026 12:19:27 GMT Cato DEM gains Root Cause Analysis with Network Adapter Metrics and Client Traceroute to pinpoint digital-experience fault domains.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds Root Cause Analysis (RCA) to Cato DEM, correlating endpoint, network, application, and historical path evidence in a single investigation to identify the likely fault domain.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds Network Adapter Metrics to the RCA workflow, surfacing adapter type, BSSID, channel, noise, and signal strength to assess local Wi-Fi conditions contributing to poor experience. • Adds Client Traceroute to the RCA workflow, providing historical hop-by-hop path evidence from the client to capture latency, loss, and path behavior — especially useful for intermittent and remote-user issues. • Extends RCA scope beyond the individual user to correlate experience data across multiple users, revealing whether an issue is isolated or shared and its broader blast radius. • Supports RCA across Cato SASE and supported third-party network environments, without requiring replacement of existing network infrastructure. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/cato-networks https://www.catonetworks.com/blog/securing-what-your-bedrock-agentcore-agents-actually-do/ https://www.catonetworks.com/blog/securing-what-your-bedrock-agentcore-agents-actually-do/ Thu, 24 Sep 2026 08:12:52 GMT Cato AI Security for Agents adds read-only discovery and inline inspection of Amazon Bedrock AgentCore agents across prompt, tool call, tool response, and output stages.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds read-only discovery of every agent running on Amazon Bedrock AgentCore, surfacing each agent's instructions, model, and connected tools without installing anything.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds inline inspection of agent-model exchanges at four observable stages — prompt, tool call, tool response, and output — to detect prompt injection and other unwanted activity before the model acts. • Extends Cato AI Security for Agents coverage to Amazon Bedrock (Agents Classic and AgentCore), Microsoft Foundry, Microsoft Copilot Studio, ChatGPT (GPTs and Agents), Amazon Quick, and developer-hosted agent endpoints. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/cato-networks https://www.catonetworks.com/blog/cato-managed-sase-for-smb/ https://www.catonetworks.com/blog/cato-managed-sase-for-smb/ Tue, 01 Sep 2026 12:21:39 GMT Cato launches SMB FlexPool and MSASE Business Center for MSPs to pool licenses and self-provision SMB customers. • New `Cato SMB FlexPool` provides MSPs and service providers a partner-level pool of Cato licenses, enabling unlimited portability to allocate and reallocate capacity across eligible SMB customers without per-customer license purchases. • New `Cato MSASE Business Center` adds a consolidated `Business Overview` page showing business accounts, orders, license usage, invoices, and customer lifecycle activity across the partner portfolio. • Authorized partners can self-provision eligible services and activate new customer accounts directly in `Cato MSASE Business Center` without waiting for manual approval workflows. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/cato-networks https://www.catonetworks.com/blog/how-cato-ai-security-keeps-up-with-claude/ https://www.catonetworks.com/blog/how-cato-ai-security-keeps-up-with-claude/ Thu, 13 Aug 2026 17:14:22 GMT Cato AI Security adds Claude inference hook integration, MCP/skill posture visibility, and scoped agent policy rollout via Cato Scout • Adds `scout-debug` command to verify hook registration inside supported agents, enabling in-workflow deployment troubleshooting. • Extends Cato AI Security coverage to Claude Chat, Claude Code, and Claude Cowork via Anthropic Inference Hooks — no endpoint agent or network inline access required. • Adds MCP and tool visibility so security teams can enumerate the MCP servers and tools available to agents and assess their combined risk surface. • Adds skill visibility and posture assessment across the environment, helping teams identify skills that require review before use in business workflows. • Enables full-lifecycle policy enforcement across user prompt, AI response, tool input, and tool output for agentic Claude interactions. • Supports scoped rollout by applying controls to specific users or groups, with Cato Scout client identity tying policies and audit records to the correct endpoint user or device. • Enables pre-emptive hook configuration — write hook policy before a supported agent is detected, so the first session on a new device is not outside enforcement. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/cato-networks https://www.catonetworks.com/blog/defeating-agentic-attacker-agentic-threat-prevention-in-action/ https://www.catonetworks.com/blog/defeating-agentic-attacker-agentic-threat-prevention-in-action/ Mon, 10 Aug 2026 09:51:35 GMT Cato Networks introduces Agentic Threat Prevention, combining inline conditional enforcement and an agentic decision-making layer to stop multi-stage AI-driven attacks in real time. • New Agentic Threat Prevention capability surfaces findings, supporting evidence, activated controls, and mitigation status inside the Cato Management Application. • Inline conditional enforcement dynamically tags hosts based on recent activity and denies specific behaviors mapped to MITRE ATT&CK tactics — for example, blocking executable downloads from cloud services on hosts that have contacted known C2 servers. • Agentic decision-making layer continuously aggregates telemetry, reasons about host patterns, and automatically applies prevention policies in real time without manual intervention — for example, denying lateral movement vectors on hosts exhibiting both internal discovery and suspicious external communications. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/cato-networks