Agent Sandboxesagentsandboxes.org
Sandbox profile · Open-source project

sandboxd

Answer in brief

MIT self-hosted engine that runs each app or agent in its own hardened Docker container on one server, with built-in coding agents, preview URLs, idle stop and wake, and per-task git checkpoints. A self-hosted engine for app-builder and agent platforms that runs each workload in a hardened container on one server; its own docs call the default container a weak security boundary and point to gVisor or separate hosts for untrusted tenants.

By the Agent Sandboxes editors · Updated

Overview

sandboxd describes itself as "the open-source engine behind AI app builders". One Go control plane drives Docker on a single host, Traefik routes preview URLs and SQLite holds state, with a reconciler that converges Docker to the database on boot. Each app gets one sandbox: a hardened runc container with its own workspace, an in-sandbox supervisor (runtimed) and a live preview URL. Coding agents run inside the sandbox (the README lists OpenCode, Claude Code, Codex and Pi; OpenCode works keyless on a free tier), and a credential-injecting proxy keeps model keys out of the container. Before every agent task the workspace is checkpointed to a private git ref, so a task can be reverted. Everything is exposed through an authenticated /v1 REST API with SSE task streams, file read and write, exec and git endpoints; the web console is an optional API client.

The Security page states that "sandboxd runs AI-generated code you did not write" and that the default shared-kernel container is "a strong isolation boundary and a weak security boundary". gVisor is available as an opt-in runtime (the vendor puts syscall-heavy work at about 1.7–4× slower), and Firecracker or Cloud Hypervisor microVMs are on the roadmap. Egress is open by default. Idle sandboxes are stopped after 35 minutes and woken on the next request, which keeps files but not memory. The project is pre-1.0 (v0.3.x) and single-server; multi-host scheduling is listed under "Later". Self-hosting is free; the vendor sells a Managed service on your own server (BYOC) and a Cloud tunnel that gives your server a public URL, and does not host compute.

Notable

  • v0.3.26 released 31 October 2026; 997 GitHub stars at verification; the README labels it "Beta · 0.x".
  • Configuration docs: /v1 auth is on by default (console session or bearer API key, no loopback bypass) and the API binds to 127.0.0.1:9090.
  • Agent tasks are checkpointed to refs/sandboxd/checkpoints/<task> before they run and can be reverted with one API call.

sandboxd specifications

AttributeDetails
IsolationOne hardened runc Docker container per app (--cap-drop=ALL, no-new-privileges, read-only root filesystem with tmpfs /tmp, memory, PID and file-descriptor limits, non-root user); gVisor (runsc) can be switched on for every sandbox with SANDBOXD_RUNTIME=runsc
DeploymentOpen-source project · BYOC available · Self-hostable
Open sourceYes (MIT)
SDKs and interfacesREST API
Cold start (vendor claim)Not documented
Max session lengthNot documented
PersistenceEach sandbox's workspace is a bind-mounted directory on the host that survives stop and reboot; the container's writable layer is read-only and /tmp is tmpfs. Idle sandboxes are stopped with docker stop after 35 minutes by default and started again on the next request, so files persist but memory and running processes are not kept. Snapshot, fork and restore endpoints exist but are labelled experimental on directory storage
GPUNot documented
NetworkingEach sandbox's dev server gets a preview URL (s-<id>-<port>.preview.<domain>) routed by Traefik; previews are public and plain HTTP by default, with TLS as a config switch. Egress is open by default in the self-hosted build: sandboxes sit on a Docker bridge and can reach the internet, the LAN and cloud metadata endpoints; an nftables egress subsystem is compiled off and a per-domain allowlist is on the roadmap. Model credentials are injected by a control-plane proxy so they never enter the sandbox
PricingFree to self-host under MIT. Managed (sandboxd installs and runs it on your own server): $199 one-time setup and $79/month ($49/month for the first 10 founding customers). sandboxd Cloud (a public HTTPS URL for a server you run): $9.99 per server per month, $4.99 with the 50% launch discount. Compute is always your own server
Free tierFree to self-host
ComplianceNot documented
RegionsNot documented
Company headquartersNot documented

Facts from sandboxd’s public documentation, verified November 18, 2026. Performance numbers are vendor claims.

Strengths and trade-offs

Strengths

  • You want a self-hosted, MIT-licensed control plane for many sandboxes with preview URLs on one ordinary server.
  • You need coding agents inside each sandbox without putting model keys in the container.
  • You want per-task checkpoints and one-call revert of an agent's changes.
  • You want idle sandboxes to stop and wake on the next request so one host holds many apps.

Trade-offs

  • The default boundary is a shared-kernel container; the vendor recommends gVisor for less-trusted code and one host per trust domain for mutually distrusting tenants.
  • Egress is open by default; a per-sandbox domain allowlist is still on the roadmap.
  • Single Docker host only; memory is not kept when an idle sandbox stops, and snapshots are experimental on directory storage.
  • Pre-1.0: minor versions may carry breaking changes.

Good fit for: Running many agent-built apps or agent workspaces on one server you control, each in its own container with a live preview URL.

How sandboxd compares

SandboxCompany HQIsolationOpen sourceSelf-host / BYOCCold start (vendor)Max sessionSnapshots / pauseGPUSDKs
sandboxdNot documentedHardened runc container (gVisor opt-in)YesYes / YesNot documentedNot documentedYes / NoNot documentedREST API
OpenSandboxNot documentedContainers or FirecrackerYesYes / No64 ms P50 (claimed)Not documentedNot documented / YesNot documentedPython, JavaScript/TypeScript, Java, Kotlin, Go, C#, CLI, MCP server, REST API
MicrosandboxNot documentedMicroVM (libkrun)YesYes / Yes<100 msNot documentedYes / Not documentedNot documentedTypeScript, Python, Rust, Go, Ruby, CLI, MCP server
DaytonaUnited StatesContainers or VMsNoNot documented / Yes<90 msUnlimitedYes / YesYesPython, TypeScript, Ruby, Go, Java
CoderUnited StatesTemplate-defined (VM/pod/container)YesYes / NoNot documentedNot documentedNot documented / YesNot documentedCLI, REST API, MCP server
OpenHands RuntimeUnited StatesDocker containerYesYes / Not documentedNot documentedNot documentedNot documented / Not documentedNot documentedPython, JavaScript/TypeScript, REST API
gVisorUnited StatesUser-space kernelYesYes / Noms (claim)Not documentedYes / Not documentedYesCLI

Head-to-head: OpenSandbox vs sandboxd

See all sandboxd alternatives →

Frequently asked questions

What is sandboxd?
MIT self-hosted engine that runs each app or agent in its own hardened Docker container on one server, with built-in coding agents, preview URLs, idle stop and wake, and per-task git checkpoints. A self-hosted engine for app-builder and agent platforms that runs each workload in a hardened container on one server; its own docs call the default container a weak security boundary and point to gVisor or separate hosts for untrusted tenants.
How does sandboxd isolate code?
sandboxd documents its isolation as: One hardened runc Docker container per app (--cap-drop=ALL, no-new-privileges, read-only root filesystem with tmpfs /tmp, memory, PID and file-descriptor limits, non-root user); gVisor (runsc) can be switched on for every sandbox with SANDBOXD_RUNTIME=runsc.
Is sandboxd open source?
Yes (MIT). It can be self-hosted.
How much does sandboxd cost?
Free to self-host under MIT. Managed (sandboxd installs and runs it on your own server): $199 one-time setup and $79/month ($49/month for the first 10 founding customers). sandboxd Cloud (a public HTTPS URL for a server you run): $9.99 per server per month, $4.99 with the 50% launch discount. Compute is always your own server Check the official pricing page for current rates.
Does sandboxd support snapshots or pause and resume?
Snapshots: yes. Pause and resume: none documented. Each sandbox's workspace is a bind-mounted directory on the host that survives stop and reboot; the container's writable layer is read-only and /tmp is tmpfs. Idle sandboxes are stopped with docker stop after 35 minutes by default and started again on the next request, so files persist but memory and running processes are not kept. Snapshot, fork and restore endpoints exist but are labelled experimental on directory storage.
What are the alternatives to sandboxd?
Common alternatives include OpenSandbox, Microsandbox, Daytona, Coder, OpenHands Runtime, gVisor. See sandboxd alternatives for a full comparison.

Sources

  1. sandboxd.io
  2. sandboxd.io/what-is-sandboxd
  3. sandboxd.io/concepts
  4. sandboxd.io/security
  5. sandboxd.io/reference/architecture
  6. sandboxd.io/reference/configuration
  7. sandboxd.io/reference/api/
  8. sandboxd.io/guides/agents
  9. sandboxd.io/guides/tasks
  10. sandboxd.io/guides/runtime-and-store
  11. sandboxd.io/roadmap
  12. sandboxd.io/managed
  13. sandboxd.io/cloud
  14. sandboxd.io/privacy
  15. github.com/tastyeffectco/sandboxd
  16. github.com/tastyeffectco/sandboxd/releases

Check before you rely on it. Products change quickly, and this page reflects public documentation as of November 18, 2026. Confirm anything that matters in the provider’s own documentation.

Cite this page

Agent Sandboxes. (2026). sandboxd. agentsandboxes.org. Last reviewed November 18, 2026. https://agentsandboxes.org/sandboxes/sandboxd/