NVIDIA OpenShell

NVIDIA/OpenShell

Open-source policy-driven sandbox runtime for autonomous AI agents, announced at GTC 2026. Enforces security constraints at the kernel level via Landlock LSM (filesystem), seccomp BPF (syscalls), and an OPA/Rego-evaluated HTTP CONNECT proxy (network) — constraints are enforced on the environment itself, so even a compromised agent cannot override them. Supports Claude Code, Codex, Cursor, and Open

Rust Apache-2.0

View on GitHub → Find it in the atlas

Install

npx skills add NVIDIA/OpenShell

Repository reader

Explore before you clone

Read the project’s README, skills and supporting Markdown here. Rendered view is shown first; source is one click away when you need the exact text.

Fetched from GitHub
Open on GitHub
Loading the project’s README…

Public files are loaded on demand from the repository. The atlas does not copy or modify them.

Screenshot

Screenshot or social card for NVIDIA OpenShell
The project's own screenshot, fetched from source rather than copied into this repository.

Platform support

Derived from the project's own README, install route, CI configuration and release assets — not from running it.
PlatformRuns hereOn what evidence
WindowsLikelyInferred from the language or runtime. Nothing in the project says so.
WSL2YesStated in the project's own README, install route, CI config or release assets.
macOSYesStated in the project's own README, install route, CI config or release assets.
LinuxYesStated in the project's own README, install route, CI config or release assets.
DockerNoNo evidence of support anywhere in the project.

Where it came from

Named by 1 of the 39 lists the atlas merges.

How it is classified

One topic, 4 harnesses. Each link is a prerendered page of everything else in that slice.

Sandbox, Security & GovernanceClaude CodeClaude / AnthropicCodex / OpenAICursor

Related in Sandbox, Security & Governance

Closest first — projects that plug into the same harnesses, then the ones the most lists agreed on.

  • Node9node9-ai/node9-proxySecurity and governance layer that checks every tool call from coding agents and MCP servers before it runs: credential jail, secret and PII…
  • AI Agent Safety Starter Packel-zachariah/ai-agent-safety-starter-packFree repo preflight scanner for Claude Code, Codex, Cursor, and other AI coding-agent runs that flags agent instructions, MCP config, risky shell…
  • gate.catBGMLAI/gate.catDeterministic, fail-closed action veto for AI coding agents that blocks irreversible shell commands (rm -rf, DROP TABLE, terraform destroy) and…
  • whatileakedselan-ai/whatileakedLocal scanner for credentials a coding agent has already written to disk: Claude Code, Codex and Cursor transcripts plus instruction/memory files…
  • Brood Boxstacklok/brood-boxRun AI coding agents (Claude Code, Codex, OpenCode) inside hardware-isolated microVMs with snapshot isolation, egress control, and MCP authorization
  • SkilLockskills-lock/skil-lockBehavior-pinning lockfile + capability-delta PR review for Claude Code & Codex skills; SARIF output for Code Scanning
  • Agent GuardJeongJaeSoon/agent-guardby JeongJaeSoon - Real-time secret-leak guardrails for AI coding agents (Claude Code, Codex), Git hooks, and CI
  • aicontainerstefanoginella/aicontainerby stefanoginella - Sandboxed devcontainer for running Claude Code, Codex, and OpenCode in bypass / auto-approve mode