Customer Event | Built In, Not Bolted On: AI Agents for Security Operations | Register Now
Master AI governance and agent control in the Netskope Community. Ask questions, browse guides, and collaborate on Skylight AI Security deployments
Recently active
Hi Everyone,I'm currently exploring the Netskope Workload Sensor functionality for AI visibility through the AI Command Center and would like to validate my understanding of the deployment architecture.Based on the documentation, my understanding of the deployment process is as follows:1. VPE Deployment First, deploy a Netskope Virtual Private Edge (VPE) in the enviornment. Register the VPE with the Netskope tenant. 2. Configure Workload Collector on VPE Navigate to Settings → Security Cloud Platform → Virtual Private Edge. Select the required VPE node/group. Under the Services tab, add and enable the Workload (eBPF) Collector. 3. Workload Sensor Deployment For Linux VMs (Ubuntu/RHEL), install the Workload Sensor as a systemd service. For Kubernetes clusters, deploy the Workload Sensor as a DaemonSet. Generate the installation command using the Netskope REST API v2 4. Communication FlowMy understanding is that the Sensor monitors process aWorkload ctivity, network connections
Looking for input from security practitioners on controlling modern AI agents like Instinct.If employees can interact with an AI agent through WhatsApp or similar messaging apps on mobile or laptops, traditional URL blocking may not be enough.How are organizations preventing or detecting corporate data exposure in this scenario?The key question: How do we prevent corporate data from reaching an AI agent when the interaction channel itself, like WhatsApp, is legitimate and allowed but the chat based AI agent like Instinct sitting inside Whatsapp chat should only be restricted to personal usage not touching corporate data on mobile\ Corp Endpoints ?
Hi everyone,If I understand correctly, AI Gateway currently supports Anthropic, OpenAI, and Google models running locally on infrastructure (thanks to the ESXi integration).Is there a solution to intercept traffic from a wider range of models—such as Qwen, Llama, DeepSeek, or others—running locally in our data center on a vLLM or LM Studio instance?Thanks !
Claude Desktop's third-party inference mode (generally available as of July 9, 2026) lets you point the app at your own cloud inference provider instead of Anthropic's first-party API - such as Amazon Bedrock, Google Cloud's Agent Platform, Microsoft Foundry, or any compatible gateway you operate. That's a big deal for regulated or security-conscious organizations: you get the full Claude Desktop experience (Chat, Cowork, and Code tabs, file creation, multi-step research) while inference and billing run through infrastructure you already control.What it doesn't give you out of the box is visibility and control over what's actually going over the wire - which prompts contain sensitive data, whether someone's about to leak a secret into a conversation, or whether a jailbreak attempt just landed. That's exactly the gap Netskope AI Gateway is built to close - sitting in front of your chosen inference provider as the policy and visibility layer that direct inference access doesn't give you.
As teams scale AI agents beyond basic chatbots, machine-to-machine traffic becomes much harder to govern.This recording covers how AI Gateway handles that agentic traffic, including a walkthrough of real-time token inspection, DLP guardrails, and model routing, plus a live demo of policy setup inside Netskope.If you're evaluating AI Gateway for your own environment, the session is a solid starting point before requesting a technical review or Proof of Value sandbox.▶ Watch here: https://netskopeacademy.skilljar.com/csg-customer-webinar-ai-gatewayOriginally shared in Community News & Announcements —
This session tackles a gap most security teams are only now running into: conventional testing tools weren't designed for LLMs, so a chatbot can pass every check and still leak data on an edge case.The recording covers why standard scans and manual pen tests fall short for AI systems, how continuous adversarial testing replaces slower manual audits, how teams are folding this testing into their AI development workflows, and how to use a Proof of Value sandbox to safely stress-test proprietary models.▶ Watch here: https://netskopeacademy.skilljar.com/csg-customer-webinar-netskope-ai-red-teaming Originally shared in Community News & Announcements —
Watch this recording to see how Netskope helps keep AI tools safe for everyday use. It checks conversations with AI as they happen, catching people trying to trick the AI into saying something it shouldn't, and blocking harmful or private information from slipping through.The session starts by explaining the risks in simple terms, then shows how the protection actually works. Good watch whether you're new to this or already using it.🎥 Watch here: https://netskopeacademy.skilljar.com/csg-customer-webinar-ai-guardrails💬 Questions after watching? Drop them below and the team will get back to you. Originally shared in Community News & Announcements —
Already have an account? Login
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.