This runs OpenClaw validation jobs on remote Linux boxes instead of your laptop, which matters when you need CI parity, Docker lanes, or just don't want to spin up your fan for the full pnpm suite. The wrapper supports both direct AWS Crabbox leases and delegated Blacksmith Testbox runs, and the docs are careful about naming them correctly (cbx_ versus tbx_ IDs). You get reusable warm boxes, full resync flags when things smell stale, and static SSH target support for macOS and Windows hosts. The real value is keeping heavy integration work off local and getting actual provider proof with lease IDs you can trace.

skills.sh installs
696
GitHub stars
1.1K
last updated
Jun 11, 2026

Install

$ npx -y skills add openclaw/agent-skills --skill crabbox --agent claude-code

Installs into .claude/skills of the current project.

Files

Crabbox

  • PR CI = default broad proof. Use the consumer repository's proof policy for required checks, builds, and ready/merge gates. Run bounded local checks only when source trust and host qualification permit them.
  • Required build unavailable: remote capacity unavailable + local execution unqualified permits an already-authorized draft with the exact missing build and capacity gap recorded. Require successful exact-head CI covering that build before ready or merge. A skipped draft job is not proof: use the consumer's supported exact-head CI route. This changes proof timing, never publication authority, source trust, credentials, owner approval, or merge gates.
  • Box only for gaps in CI/local coverage: another OS/device, live providers/channels, long E2E, heavy benchmarks. Acquire only with command ready + immediate grant.
  • Queued: stop task-owned request; use PR CI or a qualified, trusted local scope. Report uncovered requirements. Never wait on queues or chain providers.
  • Fix real failures; never rerun/re-push merely for green. Read available CI failure classification before attributing failure to change.

Repository Contract

Sponsors

Details

Repository
github.com/openclaw/agent-skills
Category
AI & Agent Building
License
MIT
First seen
Jun 11, 2026

More from openclaw/agent-skills

All 5
1aj-geddes avatare2e-testing-automationaj-geddes/useful-ai-promptsBuild end-to-end automated tests that simulate real user interactions across the full application stack. Use for E2E test, Selenium, Cypress, Playwright, browse695skills.sh installs
2dash0hq avatarotel-collectordash0hq/agent-skillsExpert guidance for configuring and deploying the OpenTelemetry Collector. Use when setting up a Collector pipeline, configuring receivers, exporters, or proces694skills.sh installs
3jwynia avatardialoguejwynia/agent-skillsDiagnose flat dialogue, same-voice characters, and lack of subtext. Use when conversations feel wooden, characters sound alike, or dialogue only does one thing 692skills.sh installs
4jwynia avatarnovel-revisionjwynia/agent-skillsManage multi-level novel revisions while preventing cascade problems. Use when editing novels, when changes at one level break things at others, when you need s691skills.sh installs
5supabase avatarskill-creatorsupabase/agent-skillsGuide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabil691skills.sh installs
6rstackjs avatarrspack-best-practicesrstackjs/agent-skillsRspack best practices for config, CLI workflow, type checking, CSS, bundle optimization, assets and profiling. Use when writing, reviewing, or troubleshooting R689skills.sh installs