sandboxing repositories

Trending repositories tagged sandboxing, ranked by trending score. Updated daily.

  1. 1
    agent-substrate/substrate

    Agent Substrate: the core system

    GitHub · Go · 7,888 stars · 865 forks · +525 stars in 24 hours · Trending score 5.04

  2. 2
    pydantic/monty

    A secure Python sandbox, written in Rust, for code written by AI.

    GitHub · Rust · 8,973 stars · 506 forks · +22 stars in 24 hours · Trending score 2.78

  3. 3
    lemonadesaltbagel/crab

    CRAB (Clean Rebuild of Anything's Behavior): rebuild software from behavior, not source code. An MCP-powered toolkit for black-box observation, clean-room reimplementation, and behavioral verification.

    GitHub · Python · 399 stars · 71 forks · +14 stars in 24 hours · Trending score 2.62

  4. 4
    NVIDIA/NemoClaw

    Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference

    GitHub · TypeScript · 23,055 stars · 3,225 forks · +5 stars in 24 hours · Trending score 2.11

  5. 5
    strands-agents/box

    Run AI agents in a sandbox that restricts what they can execute, read, write, and reach on the network. Box combines OS isolation with default-deny Dogwood policies and credential injection that keeps secrets outside the agent. Written in Rust. Supports macOS on Apple silicon and Linux.

    GitHub · Rust · 558 stars · 60 forks · +6 stars in 24 hours · Trending score 1.99

  6. 6
    brekkylab/ailoy

    The open-source agent SDK that gives every agent its own VM. Each VM is a microVM sandbox where the agent installs software, runs code and uses the GPU. Python, Node.js and Rust.

    GitHub · Rust · 158 stars · 11 forks · +2 stars in 24 hours · Trending score 1.00

  7. 7
    inoio/agents-sandbox

    Sandbox for AI coding agents (Claude Code, opencode, pi): runs each project in an isolated microVM with host-side secret injection. Linux/KVM & macOS.

    GitHub · Go · 91 stars · 10 forks · +2 stars in 24 hours · Trending score 0.98

  8. 8
    fhiltscher/awesome-ai-coding-sandboxes

    Security-posture-first list of sandboxes for running AI coding agents' code, ranked by isolation, egress control and secrets handling

    GitHub · Python · 30 stars · 22 forks · +1 stars in 24 hours · Trending score 0.91

  9. 9
    laurentftech/pi-outpost

    A self-hosted web interface for the pi coding agent: chat, files, documents and an integrated terminal, with the agent's shell confined by OS sandboxes.

    GitHub · TypeScript · 31 stars · 3 forks · +2 stars in 24 hours · Trending score 0.87

  10. 10
    tinylabscom/mvm

    Manage secure microVMs — one command from zero to secure sessions, locally and remotely.

    GitHub · Rust · 49 stars · 6 forks · +2 stars in 24 hours · Trending score 0.82

  11. 11
    getdruks/druks

    An operating system for companies: AI agents build and run software tailored to your company. Own your software instead of paying for services you barely use. Agents work in sandboxes, a person approves the important steps, and every run is recorded.

    GitHub · Python · 23 stars · 5 forks · +1 stars in 24 hours · Trending score 0.72

  12. 12
    Godcreatebugs/raytrace-cli

    Sandbox to create a local isolation for agentic run. "Eye of Sauron" for agents, observe everything from file access to network calls with rayTrace sandbox

    GitHub · JavaScript · 25 stars · 27 forks · 0 stars in 24 hours · Trending score 0.70

  13. 13
    AdvDebug/Brovan

    Run Windows and Linux programs on Windows, Linux and Android. A user-mode NT and Linux kernel emulator with WHP/KVM acceleration.

    GitHub · C# · 174 stars · 10 forks · +1 stars in 24 hours · Trending score 0.63

  14. 14
    multikernel/sandlock

    The lightest sandbox. A process-based sandbox for Linux, no container, no VM, no privilege, no prompt injection

    GitHub · Rust · 621 stars · 60 forks · 0 stars in 24 hours · Trending score 0.63

  15. 15
    Sanne/incus-spawn

    Give your AI coding agents their own machines — not your credentials. Isolated Linux workstations with full autonomy, on Linux and macOS.

    GitHub · Java · 55 stars · 31 forks · 0 stars in 24 hours · Trending score 0.57

  16. 16
    christopherkarani/rv

    rv blocks destructive shell commands from autonomous agents — 95 rule packs, on-device Laya-model judge, secrets & MCP governance, tamper-evident audit. Fail-closed. Never asks the agent.

    GitHub · Swift · 8 stars · 0 forks · +1 stars in 24 hours · Trending score 0.56

  17. 17
    brig-sh/brig

    Brig: run coding agents in a microVM sandbox - the Go cli (brig) and session daemon (brigd)

    GitHub · Go · 270 stars · 25 forks · 0 stars in 24 hours · Trending score 0.51

  18. 18
    ProjectRecon/awesome-ai-agents-security

    A living map of the AI agent security ecosystem.

    GitHub · 81 stars · 162 forks · 0 stars in 24 hours · Trending score 0.51

  19. 19
    fwdai/fletch

    A new kind of IDE for agentic engineering

    GitHub · Rust · 35 stars · 3 forks · 0 stars in 24 hours · Trending score 0.33

  20. 20
    DeepLink-org/pvisor

    Run a coding agent in a staged workspace, review its file changes, and apply only what you keep.

    GitHub · Rust · 29 stars · 8 forks · 0 stars in 24 hours · Trending score 0.33

  21. 21
    brekkylab/virtx

    Disposable Linux VMs for your own code. Runs microVMs on Linux, macOS and Windows hosts with no Docker or daemon, GPU through Vulkan, and host folders mounted in. Python, Node.js and Rust.

    GitHub · Rust · 10 stars · 5 forks · 0 stars in 24 hours · Trending score 0.31

  22. 22
    adolfintel/TDF

    Linux Gaming without Steam. Install, configure, run and distribute games inside sandboxed Wine environments

    GitHub · Shell · 36 stars · 2 forks · 0 stars in 24 hours · Trending score 0.29

  23. 23
    trailofbits/build-wrap

    Help protect against malicious build scripts

    GitHub · Rust · 43 stars · 5 forks · 0 stars in 24 hours · Trending score 0.28

  24. 24
    Dicklesworthstone/franken_engine

    Native Rust runtime for adversarial JavaScript/TypeScript extension workloads: native-only core execution (no V8/JSC/QuickJS), deterministic replay of high-impact decisions, and Ed25519-signed evidence for every containment action. Research-grade; automation ships advisory-only.

    GitHub · Rust · 35 stars · 4 forks · 0 stars in 24 hours · Trending score 0.22