Crabbox is a remote execution control plane that lets a local command run on managed cloud capacity, an SSH host, or another delegated sandbox while syncing code, streaming output, and recording evidence. It is for maintainers, contributors, teams, AI agents, and automation that need repeatable remote testing, builds, browser checks, or platform-specific validation. The catalogue add-ons provide commands and instructions for controlling Crabbox runs.
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.
npx agentmods add commands/openclaw/crabbox/capacitygit clone --depth 1 https://github.com/openclaw/crabboxWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/openclaw/crabbox/capacity)<a href="https://agentmods.dev/commands/openclaw/crabbox/capacity"><img src="https://agentmods.dev/badge/commands/openclaw/crabbox/capacity.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00741 |
| Opus 5 | $0.00000 | $0.00370 |
| Sonnet 5 | $0.00000 | $0.00148 |
| Haiku 4.5 | $0.00000 | $0.00074 |
Grade A, and why
capacity 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.
What it actually says
capacity
crabbox capacity [--json] reports the authenticated owner's current admission
count and effective owner limit. It requires a configured coordinator and uses
the normal coordinator credentials, including the existing resolved shared or
unknown owner identity. It never falls back to a direct provider or local lease
list. Older coordinators without GET /v1/capacity return an unsupported error.
crabbox capacity
crabbox capacity --json
The only option is --json (plus --help). There are no positional arguments or
owner, user, org, month, or scope selectors. Even admin requests report only the
owner resolved by normal authentication. The API rejects every query parameter
with HTTP 400.
self-owner admission count: owner=github:12345 activeLeases=10
effective owner limit: 10
observed at: 2026-09-02T12:00:00.000Z
Snapshot only; not a reservation or approval to allocate.
JSON contains exactly four fields:
{
"owner": "github:12345",
"activeLeases": 10,
"effectiveLimit": 10,
"observedAt": "2026-09-02T12:00:00.000Z"
}
effectiveLimit: 0 means the selected owner limit is off; text prints off.
The selected limit uses the existing capacity-owner policy, independently of
admin authentication. A positive elevated limit for a configured member selects
the larger of ordinary and elevated limits, including when the ordinary limit
is zero. The response does not expose membership or the underlying limit config.
activeLeases counts existing admission entries only, with no candidate added.
At ten existing leases and a limit of ten, the diagnostic succeeds with 10;
an attempted new allocation would instead be checked as 11/10. Successful
diagnostics exit 0 even at or above the cap. Invalid syntax/configuration and
request errors follow normal CLI error handling; errors go to stderr and data
go to stdout. Malformed successful responses fail instead of displaying zeros.
Unlike monthly usage, the count spans every creation month and org for the exact, case-sensitive owner identity. Active and provisioning managed leases count even after their canonical expiry, until a terminal state is committed. Released, expired, failed, and registered inventory records do not count. A live, unexpired admission reservation overrides its canonical lease by ID and counts once; reservation-only IDs count too. Stale reservations are ignored without cleanup, with canonical records used when present. Shared leases owned by someone else and direct-only leases absent from the coordinator do not count.
This read-only snapshot uses the same serialized boundary as admission on Cloudflare and Node. Normal service startup and background recovery retain their existing behavior; the capacity handler itself neither allocates nor reconciles resources, deletes stale records, schedules maintenance, nor calls providers. It reveals no lease IDs, records, org/provider breakdowns, costs, or other owners' counts. The aggregate is a narrow exception to owner/org visibility, not expanded access to lease records or monthly reports.
The snapshot is not a reservation or approval to allocate. Fleet, org, budget,
and provider gates may still reject allocation, and the count can change after
observedAt. See cost and usage and
auth and routing.
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.
- 2d ago First seen · 73 lines · 0 tokens per session scan A 4db49b97b523
capacity is a command published in the GitHub repository openclaw/crabbox (1,368 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 741 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 2026-09-03.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.