OpenParable
Skip to content
← Back to skills

Portaljs Deploy

DevOps & Security
portaljs-deploy

Portaljs Deploy is an Agent Skill for devops & security published in the datopian/portaljs repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Deploy a PortalJS portal to PortalJS Arc — Datopian-managed static hosting on Cloudflare. Builds a static export, uploads it, and returns a live SLUG.arc.portaljs.com URL. One command, one target. Use when a portal is ready to publish or redeploy to a live URL.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/datopian/portaljs/blob/HEAD/skills/portaljs-deploy/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/portaljs-deploy, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

PortalJS — Deploy

Overview

Publish an existing PortalJS portal to PortalJS Arc — Datopian's managed static hosting on Cloudflare. Build a static export, upload it to the Arc API, and print a live https://SLUG.arc.portaljs.com URL. Re-running redeploys the same portal (idempotent on the slug). This is a single-target skill — it deploys to Arc only. For self-hosting, run npm run build and upload out/ to any static host; no skill required for that path. Arc serves static exports only — SSR is not hosted on Arc yet.

Prerequisites

  • A PortalJS portal directory with a package.json that lists next as a dependency.
  • Node 18+ and npm on PATH (the Arc device-login flow uses Node's global fetch).
  • curl and tar available for packaging and upload.
  • A PortalJS Arc token — read from PORTALJS_TOKEN, or ~/.portaljs/credentials ({"token":"…"}). If neither exists, the skill signs in on demand via a device-code flow; no manual token copying required.
  • If the portal stores large data in Git LFS/R2, do not run git lfs pull before deploying — large datasets are served from Cloudflare R2 via absolute URLs in datasets.json, not copied into the export.

Instructions

The canonical, full step-by-step workflow is .claude/commands/portaljs-deploy.md — the single source of truth. Read and follow it when executing. Summary:

  1. Gather input — portal directory (default .) and slug (default from package.json name or directory name, slugified). Confirm the directory is a Next.js project; reject reserved slugs (www, api, admin, staging, arc).
  2. Resolve the Arc token: read PORTALJS_TOKEN, else ~/.portaljs/credentials; if missing, run the device-authorization sign-in flow and save the returned token.
  3. Ensure next.config.js sets output: 'export' and images: { unoptimized: true }, then run npm run build; stop if the build fails.
  4. Verify the export carries no dataset bytes — run npm run check-export (or scripts/check-export.mjs) to catch Git LFS pointer leaks and oversized data files.
  5. Tar the out/ directory and POST it to $PORTALJS_ARC_API/v1/deploy?slug=<slug> with the bearer token; handle 200/401/409/400/413 responses distinctly.
  6. Report the live URL, file count, upload size, and R2-vs-inline dataset counts.

Output

  • Modified (if needed): next.config.js — adds output: 'export' and images: { unoptimized: true } when absent, preserving the rest of the config.
  • Created (on first sign-in): ~/.portaljs/credentials (mode 0600).
  • Verified: npm run build exits 0, out/index.html exists, the export-hygiene check passes.
  • Result: the portal is live at https://SLUG.arc.portaljs.com; re-running updates the same slug in place.

Error Handling

SymptomCauseFix
NOT_A_PORTAL errorNo next dependency found in PORTAL_DIR/package.jsonRun from a valid portal directory, or pass the correct path.
Slug rejectedDerived slug is reserved (www, api, …) or not a valid DNS labelPass an explicit --slug <name>.
Build fails (non-zero exit)App/config error surfaced in npm run buildPrint the log, fix the error, never deploy a failing build.
check-export failsGit LFS pointer leaked into out/, or a data file exceeds the size budgetReference large data by absolute R2 URL via portaljs-add-dataset; don't git lfs pull before building.
401 on uploadToken invalid, expired, or revokedRe-run the device sign-in flow once, retry the upload; stop if it 401s again.
409 on uploadSlug already taken by another accountChoose a different --slug.
400 / 413 on uploadMalformed slug or export too largeRead the JSON error field and address the specific cause.

Examples

Example 1 — Deploy the current directory with the default slug

/portaljs-deploy

Example 2 — Deploy with an explicit slug

/portaljs-deploy --slug my-open-data

Example 3 — Non-interactive deploy from CI with a token env var

export PORTALJS_TOKEN=arc_live_xxxxxxxx
/portaljs-deploy ./portals/city-budget --slug city-budget

Resources