Borrowing it
Nothing to install: this file belongs to openclaw/shellbench. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/openclaw/shellbench/main/.agents/skills/crabbox/SKILL.mdgit clone --depth 1 https://github.com/openclaw/shellbenchWrote 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/skills/openclaw/shellbench/crabbox)<a href="https://agentmods.dev/skills/openclaw/shellbench/crabbox"><img src="https://agentmods.dev/badge/skills/openclaw/shellbench/crabbox/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/openclaw/shellbench/crabbox"><img src="https://agentmods.dev/badge/skills/openclaw/shellbench/crabbox.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00036 | $0.02849 |
| Opus 5.5 | $0.00014 | $0.01140 |
| Sonnet 5.5 | $0.00007 | $0.00570 |
| Haiku 4.5 | $0.00004 | $0.00285 |
Grade A, and why
crabbox 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 2mo 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.
How it starts
The opening of the file, as written. The whole thing — 318 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Crabbox
Use Crabbox when ClawBench needs remote Linux proof for CI-parity checks, broad Python suites, package/build lanes, provider/HF secrets, hydrated agent dotfiles, benchmark sweeps, warmed reusable boxes, sync timing, logs/results, cache inspection, or lease cleanup.
Default backend: blacksmith-testbox. The separate blacksmith-testbox skill
has been removed; this skill owns both the normal Crabbox path and the direct
Blacksmith fallback playbook.
First Checks
- Run from the repo root. Crabbox sync mirrors the current checkout.
- Check the wrapper and provider support before remote work:
command -v crabbox
../crabbox/bin/crabbox --version
crabbox run --provider blacksmith-testbox --help | sed -n '1,140p'
- Prefer the sibling
openclaw/crabboxcheckout at../crabbox/bin/crabboxwhen present. The user PATH shim can be stale. - Check
.crabbox.yamlfor repo-owned AWS defaults, but override the provider explicitly for normal maintainer validation. Even if config still says AWS, broad ClawBench proof should normally pass--provider blacksmith-testbox. - Prefer local targeted tests for tight edit loops. Broad suites, package checks, Docker/container sweeps, live/provider lanes, and benchmark sweeps belong remote.
- Do not move broad gates back onto the laptop just because Testbox is queued or capacity is constrained. Keep targeted local edit-loop checks narrow, leave the remote lane queued, or report the capacity blocker.
- For live/provider bugs, check keys on the local Mac before downgrading to
mocks. Source local
~/.profileand test only presence/length. If Crabbox does not already have the key, copy only the exact needed key into the remote process environment for that one command. Do not print it, sync it as a repo file, or leave it in remote shell history or logs.
Default Blacksmith Backend
Use this for python -m pytest -q, full Ruff lint, Docker/container sweeps,
package/build checks, provider/HF lanes, or anything likely to fan out beyond a
small targeted test.
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.
- 2mo ago First seen · 318 lines · 36 tokens per session scan A 7d116d8c3998
crabbox is a skill published in the GitHub repository openclaw/shellbench (141 stars, last pushed 23d ago), licensed MIT. It adds 36 tokens to every session and 2,849 once invoked, about $0.0001 per session on Opus 5.5. 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.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.