La boutique est ouverte

Nous construisons les machines Workstation AI pour les sites web, les agents IA et l'entraînement de LLM — configurez-les et achetez-les en ligne.

Obtenir les prix pro
Produits IA
Labs IAAgents OpenAIAgents ClaudeGrok BotCRM natif Workstation AIMarketingTous les produits
Solutions IA
IA appliquéeAgents IAIA générativeAutomatisation par IAMontée en compétence IAOffres IA pour PMELaboratoire IA d'entrepriseStations de travail IAMatériel et logiciels NVIDIACalcul accéléré NVIDIAIA privéeClusters GPUIA en périphérieIA par secteur
Services IA
Modernisation des plateformesIngénierie numériqueFondations data et IAOpérations autonomesConseil IAAutomatisation DevOpsCybersécuritéDéveloppement logicielCréation d'agentsMise en place MLOpsLocation de GPU
À propos
EntrepriseNotre histoirePartenairesTémoignages clients
Centres de données IA
Analyses IASérie JevLabo de benchmarksChoix de modèles
Notes techniques
Documentation
OS LinuxUbuntuRed Hat Enterprise LinuxSUSE LinuxAlpine LinuxDebian, Rocky & AlmaLinux
Plateformes cloud d'IAAWS & Amazon BedrockGoogle Cloud & Vertex AIMicrosoft Azure & AI FoundryClouds et places de marché GPU
Centres de données IA privésPour les cabinets d'avocatsPour les cabinets comptablesPlateformes sur siteCloud entièrement sécurisé
OutilsRing PromoterJobshout
Actualités
Blog
Nous contacterSe connecter

Stations de travail IA, logiciels multi-agents IA, infrastructure GPU et solutions d'agents intelligents pour les entreprises modernes.

Nous contacter

Solutions IA

IA appliquéeAgents IAIA générativeAutomatisation par IAMontée en compétence IAOffres IA pour PMELaboratoire IA d'entrepriseIA par secteur

Produits

Tous les produitsWorkstation AI native CRM & ERPMarketingAgents OpenAIRing PromoterJobshout

Entreprise

À proposEntrepriseCentres de données IAPourquoi WorkstationPartenairesTémoignages clientsTarifsContactTenthMatrix

Ressources

Analyses IANotes techniquesDocumentationOS LinuxPlateformes cloud d'IACentres de données IA privésBlogRechercherPlan du site
Bureau du Royaume-Uni
77-79 Marlowes, Hemel Hempstead HP1 1LFItinéraire : prenez la sortie 20 de la M25, Outer LondonN° d'entreprise: 11641870Lun - Ven : 9h00 - 18h00 GMT
+44 7515 356 146
Bureau de Belgique
Workstation SRL, Rue Vanderkindere 34, 1180 Uccle, BrusselsBE 0751.518.683Lun - Ven : 9h00 - 18h00 CET
+32 492 45 67 46
Bureau de l'Inde
#159 Sector 9, Pocket 1, DDA Flats, 110077 Dwarka, New Delhi
+91 98881 98841

© 2026 Workstation AI. Tous droits réservés.

ConfidentialitéSuppression des donnéesCookiesConditions d'utilisationPlan du site web

NVIDIA, the NVIDIA logo, RTX, RTX PRO, Blackwell, CUDA, NVLink, MGX, Omniverse, Spectrum-X, Quantum, BlueField, DGX and HGX are trademarks and/or registered trademarks of NVIDIA Corporation in the US and other countries. AMD, the AMD Arrow logo, EPYC, Instinct, Ryzen and Threadripper are trademarks of Advanced Micro Devices, Inc. Dell and PowerEdge are trademarks of Dell Inc. or its subsidiaries. Intel and Xeon are trademarks of Intel Corporation or its subsidiaries. AMAX, AceleMax and LiquidMax are trademarks of AMAX Engineering Corporation. All other product names, logos and brands are the property of their respective owners. We use them only to identify the products we describe, supply or support; their use does not mean their owners endorse us.

Loading blog...

Accueil / Blogs
OpenshellNvidiaSentrySecurityAI Agents

OpenShell, Announced at GTC Berlin: Policy Outside the Agent

Huang's security film rejected 'please don't' as a control. OpenShell enforces a policy on syscalls, files and network, with Sentry on BlueField-4.

Workstation team9 novembre 20261 min read
OpenShell, Announced at GTC Berlin: Policy Outside the Agent

At GTC Berlin, Jensen Huang's security film said the only control on many agents today is asking them nicely not to delete files or touch SSH keys. OpenShell is NVIDIA's replacement: a policy, enforced outside the model, on system calls, files and network connections.

OpenShell and Sentry cover

OpenShell request path

What was announced

Each agent runs in its own sandbox with no direct network path. Outbound calls go to a supervisor, which allows or denies them. Keys stay outside the sandbox and are attached only on the way to an approved endpoint. A policy change is checked against a permission ceiling so the agent cannot be given more than intended. Sentry, optional, runs on BlueField-4 and can cut the model connection if the agent's plan leaves policy. Huang said that quarantine is in milliseconds. OpenShell is Apache 2.0, and the film said you install a CLI, create a sandbox, and run the agent you already use.

The reference design puts models on Vera Rubin, harnesses on Vera CPUs inside OpenShell, and Sentry on BlueField. Huang's published line: think on Rubin, act on Vera, governed on BlueField. The newsroom list of companies working with that design is long, and NVIDIA marks much of it as when-and-if-available.

The full path, the partner list, and which numbers are NVIDIA's are in the long article. How the CLI sits on a Mac Studio or a Linux GPU box is in our earlier OpenShell operator note.