deploy

SkillCloud & infra

Deploy any branch to its HF Space directly from the CLI (local, no push) and monitor until the Space is live.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the deploy skill

What this skill tells your AI

The instructions your AI receives, as published by qud-technologies/quranic-universal-audio in .claude/skills/deploy/SKILL.md and read by ahel’s review.

Deploy the Inspector to its Hugging Face Space straight from the current checkout — no git push, no GH-Actions runner. The deploy stages the git-tracked tree, uploads it to the Space repo, and factory-reboots the Space, so it reflects the committed state of whatever branch/worktree you run it from. Default target is the dev Space; pass prod to target prod.

Run

  1. Deploy (foreground, ~15-30s to upload + trigger the rebuild):

    bash .claude/skills/deploy/scripts/deploy.sh dev
    
  2. Monitor readiness as a background Bash job (run_in_background: true) so you're notified when the Space is actually live — it polls the HF Space runtime stage until RUNNING and /healthz returns 200, not just when the upload finished:

    python .claude/skills/deploy/scripts/wait_space.py dev
    

    Exits 0 when RUNNING + healthy, non-zero on BUILD_ERROR / RUNTIME_ERROR / a paused-or-stopped Space / timeout.

Notes

  • Local — no remote branch required. It stages the tracked files of the checkout you run it from; run it from the worktree on the branch you want live. Untracked files are not staged.
  • Targets: dev → hetchyy/quranic-inspector-dev, prod → hetchyy/quranic-universal-audio.
  • CI path instead (runs the full check suite, deploys a pushed ref): gh workflow run inspector-deploy.yml --ref <branch> -f env=dev.

Signals

GitHub stars
33
Forks
4
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
deploy-qud-technologies
Source
github.com/qud-technologies/quranic-universal-audio