checkpoint

A command for saving and restoring the state of a development machine or workspace. A checkpoint can preserve installed packages, tools, caches, generated test data, services, or a paused bug investigation for later reuse.

In plain words
What is it for?
Use it to create, inspect, list, restore, fork, delete, prune, and manage policies for saved development environments or workspace states.
Why use it?
It avoids repeating costly setup when moving work to another machine or starting a separate copy. Checkpoints can also be inspected, listed, restored, copied, removed, or cleaned up.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/openclaw/crabbox/checkpoint
Clone the repo
git clone --depth 1 https://github.com/openclaw/crabbox
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 7,607 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.07607
Opus 5 $0.00000 $0.03803
Sonnet 5 $0.00000 $0.01521
Haiku 4.5 $0.00000 $0.00761

Measured 2d ago against content hash ccd166626129, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

checkpoint scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

docs/commands/checkpoint.md · 660 lines

How it starts

The opening of the file, as written. The whole thing — 660 lines — stays where its author put it; the contents beside it link to each section on GitHub.

checkpoint

Save the state of a lease, then restore it onto another box or fork it into a fresh lease later. A checkpoint turns expensive one-time setup — installed dependencies, warmed caches, generated fixtures, a paused bug — into something you can reproduce on demand without repeating the work.

Each checkpoint has an ID of the form chk_<hex>. Newly created brokered native AWS, Azure, and GCP checkpoints are owned durably by the coordinator; their local records are recoverable caches. Direct-provider, archive, recipe, and historical checkpoints remain owned by local records under $XDG_STATE_HOME/crabbox/checkpoints, or <user-config-dir>/crabbox/state/checkpoints when XDG_STATE_HOME is unset.

Subcommands: create, list, inspect, policy, restore, fork, delete, prune.

Two checkpoint kinds

Native (provider snapshot or image) — captures the whole machine: packages, tools, caches, services. Stored in the provider account, so it incurs provider storage costs. Recorded as one of aws-ami, aws-ebs-snapshot, azure-managed-image, azure-os-disk-snapshot, gcp-machine-image, gcp-disk-snapshot, hetzner-snapshot, machine0-image, parallels-snapshot, or daytona-snapshot.

Archive (workspace tarball) — captures only the contents of the remote workdir as workspace.tar.gz. Portable across any POSIX SSH lease, but it does not preserve machine state.

--mode auto (the default) picks native for providers that support it on the current lease and falls back to an archive otherwise. There is also a metadata-only recipe kind produced by --recipe-only, which records the checkpoint without creating any artifact.

Both kinds may contain secrets. Native checkpoints capture the full root volume (caches, logs, credentials); archives capture build outputs and generated files. Delete checkpoints when you no longer need them.

Quick start

# Warm a lease, do the expensive setup, then snapshot it
crabbox warmup --provider aws --class beast
crabbox run --id swift-crab --shell 'npm ci && npm test'
crabbox checkpoint create --id swift-crab --name after-npm-ci
# checkpoint created id=chk_abc123 kind=aws-ebs-snapshot resource=snap-... state=available region=eu-west-1 workdir=...

# Fork the checkpoint into a brand new lease
crabbox checkpoint fork chk_abc123 --class beast
# checkpoint forked id=chk_abc123 lease=cbx_... slug=purple-whale image=snap-... workdir=...

# Run against the forked lease
crabbox run --id purple-whale -- npm test

Read the full file on GitHub · 660 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 660 lines · 0 tokens per session scan A ccd166626129

Subscribe to this mod's changes

checkpoint is a command published in the GitHub repository openclaw/crabbox (1,343 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 7,607 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.