Workflow orchestration / Pipekit alternative

Pipekit Argo alternative

Pipekit is a control plane for Argo Workflows. It adds multi-cluster visibility, access controls, run history, logs and a self-service workflow interface while execution stays on your Kubernetes clusters. Those responsibilities differ from providing each agent with a returning Linux computer.

Start buildingCompare capabilities ↓

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

When Pipekit fits

Pipekit fits platform teams operating Argo pipelines with governance, observability, cluster routing and an optional self-hosted control plane. Keep it when workflow scheduling and organizational access are the product requirement.

When to consider Sandbox0

Consider Sandbox0 for the durable Linux execution environment used by an agent or custom workflow step. You still need workflow definitions, scheduling, retries, organization-level governance and any Argo integration you depend on.

Capabilities

Pipekit and Sandbox0: what persists?

Feature comparison with documentation and verification status
Environment and runtime lifetimePipekitNot established in this reviewSandbox0ttl 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 filesPipekitOwn-cluster production storage is deployment-specific; the optional trial explicitly lacks persistent storage.View evidence ↗Sandbox0The writable RootFS survives successful checkpoints, including software, system configuration, dependencies, and user data.Read documentation →
Whole RootFSPipekitRun history, logs and workflow artifacts are not a standard whole writable RootFS contract.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 statePipekitStopping/retrying a workflow is not arbitrary-process checkpoint/resume.View evidence ↗Sandbox0Pause compute between tasks; resume on available capacity. Default resume starts fresh processes from the saved filesystem.Read documentation →
Memory and process statePipekitA managed process-memory snapshot interface is not established by the reviewed control-plane documentation.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.

Pipekit orchestrates Argo on your infrastructure. An artifact repository or persistent volume you configure is not a documented automatic whole-RootFS growth contract. Sandbox0 owns that runtime lifecycle; integrating a workflow control plane remains separate work.

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.

Workflow logs and archived run metadata can remain available without a worker, but that is not the same as supported access to a paused agent’s full RootFS. The reviewed Pipekit control-plane docs do not establish that file interface.

See paused file operations →

The control plane and execution cluster are separate

Both deployment shapes run a Pipekit Agent and Argo in your cluster. Cloud hosts the control plane; self-hosting puts its UI, API and database in your infrastructure and can support air-gapped use. Workflow data, metadata and logs have distinct placement policies; none alone establishes retention of an arbitrary worker’s complete RootFS.

Source 1 ↗

Trial-cluster limits are not universal production limits

The optional free trial cluster explicitly disallows persistent storage. Its in-memory artifact repository can lose artifacts at any time. The supplied workflow defaults include a one-hour active deadline and cleanup, but the guide says individual workflows can override defaults. Do not turn that trial example into a platform-wide maximum runtime or a statement about your own cluster’s storage.

Source 1 ↗

An MCP client submits workflows; it does not supply a saved computer

The current CLI manages submissions, runs, stopping, retries and schedules, and hosts an MCP server for agent clients. That lets an agent invoke a workflow control plane. The persistence and recovery of the commands inside each workflow still depend on its execution/storage design.

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.

Pipekit

Pricing reviewed 2026-10-18

Pipekit commercial service
Current numeric tariff not verified
Obtain terms for the selected Cloud or self-hosted control plane, support and execution deployment.
Rate source ↗
Optional evaluation cluster
Free during the trial period
Evaluation limits apply; this is not persistent production storage.
Rate source ↗
Production execution infrastructure
Your cluster and storage costs
Separate these from the control-plane quote, support and operational effort.
Rate source ↗

Obtain a current commercial quote; no numeric production tariff is established here. Cloud and self-hosted deployment models can have different support and license terms. The trial cluster’s limits cannot be raised, while its documented workflow defaults can be overridden per workflow.

Check current Pipekit 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 Pipekit

  1. Inventory Argo templates, cluster identities, volume/artifact policy, retries, schedules and required governance or air-gapped operation.
  2. Retain the workflow control plane where needed and integrate compatible Sandbox0 execution explicitly; export application files from their actual storage backend.
  3. Test durable system state, RootFS growth, paused files, cancellation/retry semantics and both cluster/control-plane and runtime costs.

Common questions

Is Sandbox0 a drop-in Pipekit replacement?

Evaluate API compatibility and product scope. These products serve different core workflows, so a change may require a different architecture.

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.pipekit.io/concepts/cloud-vs-self-hosted.md ↗docs.pipekit.io/get-started/free-trial-cluster.md ↗docs.pipekit.io/reference/cli.md ↗docs.pipekit.io/ ↗pipekit.io/pricing ↗Sandbox0 RootFS capacity →

Related comparisons

Expanse Cluster Intelligence vs Sandbox0 →Google Agent Runtime vs Sandbox0 →Hatchet vs Sandbox0 →Hyrex vs Sandbox0 →Mastra vs Sandbox0 →Microsoft Foundry hosted agents vs Sandbox0 →