Agent sandboxes / Islo alternative

An Islo sandbox alternative for long-lived agent environments

Islo targets persistent agent computers and captures the entire filesystem, including runtimes and system configuration. Its creation-time lifecycle policy, selected disk size, and pause behavior are the useful comparison points for an agent that returns to evolving state.

Start buildingCompare capabilities ↓

Reviewed 2026-10-18 · Published by the Sandbox0 team

When Islo fits

Islo fits coding agents, full-stack previews, and evaluation workflows that benefit from its persistent sandbox, gateway, snapshot, and CLI integration. Its policy can pause on idle or total active time and resume when new work arrives.

When to consider Sandbox0

Consider Sandbox0 when disk capacity should expand automatically within the configured RootFS ceiling and supported persistent-file access should leave compute paused. Your application can keep model and harness choices separate from the runtime.

Capabilities

Islo and Sandbox0: what persists?

Feature comparison with documentation and verification status
Environment and runtime lifetimeIsloThe guide supports unattended hours or days and allows omitted lifecycle timers; account limits still require verification.View evidence ↗Sandbox0ttl and hard_ttl default to 0 (disabled). Configure idle pause and deletion policies for the same persistent sandbox identity; an idle running sandbox still accrues compute charges.Read documentation →
Persistent filesIsloPaused sandboxes keep their filesystem for resume.View evidence ↗Sandbox0The writable RootFS survives successful checkpoints, including software, system configuration, dependencies, and user data.Read documentation →
Whole RootFSIsloSnapshots explicitly retain the entire filesystem, installed dependencies, and system configuration.View evidence ↗Sandbox0Persistence extends beyond /workspace to paths such as /usr/local, /etc, and /home. Runtime tmpfs and virtual filesystems are excluded.Read documentation →
Release compute, keep stateIsloManual or policy-driven pause and resume are available.View evidence ↗Sandbox0Pause compute between tasks; resume on available capacity. Default resume starts fresh processes from the saved filesystem.Read documentation →
Memory and process stateIsloThe reviewed snapshot contract specifies filesystem state, without an explicit running-process or RAM restoration guarantee.View evidence ↗Sandbox0Experimental memory checkpoints can preserve compatible process state. Use memory: true on both pause and resume on a compatible deployment. A filesystem checkpoint alone does not preserve running processes.Read documentation →

Online RootFS growth

On growth-enabled images, Sandbox0 automatically expands the writable RootFS as free space runs low, within a configured sparse logical ceiling up to 64 TiB. Agent processes keep running. Growth is enabled in the qualified Cloud region; confirm activation for other regions. Existing images and snapshots retain their captured storage geometry. The ceiling is logical capacity, not preallocated host disk.

Islo accepts a disk size for new sandboxes. The reviewed lifecycle and snapshot guides do not establish automatic online expansion of the mounted root filesystem when free space runs low.

Files while compute is paused

Sandbox0's Files API reads persistent files without resuming compute. Filesystem-only pauses also permit edits; retained-memory pauses are read-only and reject mutations with 409. SSH, SFTP, file watching, and runtime-only mounts require running compute.

Islo documents file access as activity that can auto-resume a paused VM. Sandbox0’s supported paused file operations do not trigger auto_resume; memory-paused state remains read-only.

See paused file operations →

Filesystem snapshots include more than the workspace

Islo’s snapshot guide explicitly captures files across the sandbox filesystem, installed tools, and system configuration. Restore creates a new sandbox. That already covers a broader state surface than a source-code folder; verify process restoration separately instead of assuming filesystem state is RAM.

Source 1 ↗

Lifecycle policy is fixed when the sandbox is created

Islo distinguishes idle pause, total-active-time pause, and deletion from creation. Omitted duration fields disable their behavior. The guide says an existing policy cannot be edited, so changing it requires creating another sandbox or restoring saved state. Review deletion separately from a reusable pause.

Source 1 ↗

File activity can wake a paused Islo sandbox

With auto_resume set to on_activity, exec, file access, and webhook work can resume compute. Sandbox0’s supported paused Files API path keeps compute paused; memory-paused mutations remain protected. Match that difference to background file inspection and control-plane workflows.

Source 1 ↗Source 2 ↗

Pricing

Compare billing models, then price your workload

CPU allocation, memory, billing minimums, disk, plan fees, and supported workload differ. These meters are not equal-performance machine quotes. Confirm region, resource units, and included allowances before estimating a total.

Islo

Pricing reviewed 2026-10-18

Running CPU
$0.07 per CPU-hour
The pricing FAQ defines the CPU meter as vCPU-hours while the computer runs.
Rate source ↗
Provisioned memory
$0.04 per GB-hour
Rate source ↗
Running disk
$0.0007 per GB-hour
The current page says CPU, memory, and disk are charged while running, with no charge when stopped.
Rate source ↗

The page lists no seat licenses or idle fees and offers startup credits. Verify separate retained-snapshot terms, account limits, and BYOC commercial terms. Its published GB memory and disk units are preserved.

Check current Islo pricing ↗

Sandbox0

$0.015/GiB-hour compute (billed by the second at approximately $0.000004167/GiB-second), plus $0.02/GiB-month of retained incremental RootFS and memory checkpoint data. Shared base images and platform recovery retention are excluded; paused sandboxes accrue storage charges only.

Compute / GiB-hour
$0.015
Retained data / GiB-month
$0.02
Paused compute
$0.00

A 4 GiB agent, 100 running hours

$6.00 compute

4 GiB × 100 hours, billed by the second. Keeping 20 GiB of billable incremental data for a full billing month adds $0.40. CPU is derived from memory by the platform; choose a tier that fits your workload.

Understand storage billing →

Before migrating from Islo

  1. Record creation-time lifecycle policy and required files, and save a snapshot before deleting an environment that needs to be retained.
  2. Port images, setup scripts, gateway credentials, network access, and session integration, keeping filesystem and process-restoration requirements separate.
  3. Test repeated tasks, packages outside the workspace, automatic RootFS growth, and file reads and edits while Sandbox0 compute remains paused.

Common questions

Is Sandbox0 a drop-in Islo replacement?

Evaluate API compatibility and product scope. A shared sandbox category does not make SDKs, lifecycle semantics, or limits interchangeable.

Does keeping files also keep a running agent?

Filesystem state and volatile process state are separate. Sandbox0 requires explicit compatible memory checkpoints to preserve running processes; its default filesystem resume starts fresh processes.

Sources and review scope

The Boat team's Battleships directory is the discovery source. References below describe the product and plan discussed on this page. Check current availability, regional terms and billing units before migrating. Where the cited material does not establish a capability or price, the comparison identifies that uncertainty.

Battleships research record ↗docs.islo.dev/cli/snapshots.md ↗docs.islo.dev/concepts/sandbox-lifecycle.md ↗/docs/sandbox/files ↗islo.dev/pricing/ ↗Sandbox0 RootFS capacity →

Related comparisons

Agent 37 vs Sandbox0 →AgentComputer vs Sandbox0 →ainclave vs Sandbox0 →Alibaba Cloud Agent Sandbox / FC / AgentRun vs Sandbox0 →Alibaba Cloud AgentBay vs Sandbox0 →Amika vs Sandbox0 →