Use Vercel Action logo

Use Vercel Action

Community
amondnet
use-vercel-action

Wire `amondnet/vercel-action` into a GitHub Actions workflow to deploy Vercel projects from CI. Use when the user asks to deploy to Vercel from GitHub Actions, add Vercel preview deploys, post a Vercel preview URL on a PR, configure prebuilt Vercel deployments, set up alias domains for previews, create a GitHub Deployment for Vercel, or migrate off Vercel's native GitHub integration. Covers required tokens/IDs, the CLI vs experimental-API modes, prebuilt flows, alias placeholders, deprecated `zeit-*`/`now-*` aliases, and the action's outputs (`preview-url`, `preview-name`, `deployment-id`).

Overview

Publisheramondnet
Repositoryvercel-action
Skill nameuse-vercel-action
Stars
764
Forks
118
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by amondnet on GitHub. Read the source before you install it.

Installation

Install the Use Vercel Action AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/amondnet/vercel-action/tree/master/skills/use-vercel-action
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/amondnet/vercel-action.git /tmp/vercel-action
mkdir -p .claude/skills
cp -r /tmp/vercel-action/skills/use-vercel-action .claude/skills/use-vercel-action
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Use Vercel Action in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Use Vercel Action on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Use Vercel Action is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

use-vercel-action

This skill teaches an agent how to consume amondnet/vercel-action correctly. Inputs and outputs documented here come from action.yml in this repo — when in doubt, open action.yml for the full surface.

When to use

  • Adding amondnet/vercel-action to a workflow for the first time.
  • Fixing a workflow that already uses the action (input names, mutual exclusions, missing prerequisites).
  • Choosing between CLI mode (default), experimental API mode, and the two prebuilt flows.
  • Migrating from Vercel's native GitHub integration to action-driven deploys.

When NOT to use

  • The user just wants the Vercel CLI locally (no GitHub Actions involved) — refer them to the official Vercel CLI docs instead.
  • The user is happy with Vercel's native GitHub integration and isn't asking for action-based deploys.

Mandatory prerequisites

Consumers consistently miss these. Verify each before writing the workflow:

  1. Disable Vercel's native auto-deploys in the project repo, otherwise both Vercel and the action will deploy and you'll get duplicate previews. Add to vercel.json (or vercel.ts):

    json
    { "git": { "deploymentEnabled": false } }

    Legacy projects may instead use the older { "github": { "enabled": false } } — both are accepted.

  2. Link the project locally and capture the IDs. Run vercel link once on a developer machine; this writes .vercel/project.json with projectId and orgId. Store those values as GitHub repository secrets:

    Do not commit raw IDs into workflow files; reference them via ${{ secrets.* }}.

  3. GitHub Deployments need explicit permission. If you set github-deployment: true, the workflow job must declare permissions: deployments: write (and usually contents: read). Without it the deployment write call fails.

  4. Action runtime is Node 24. No consumer-side Node setup is needed for the action itself — the runner provides it.

Quick start: preview deploy on every PR

yaml
name: Preview Deploy
on:
  pull_request:
    branches: [main]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: amondnet/vercel-action@v42
        with:
          vercel-token: ${{ secrets.VERCEL_TOKEN }}
          vercel-org-id: ${{ secrets.VERCEL_ORG_ID }}
          vercel-project-id: ${{ secrets.VERCEL_PROJECT_ID }}
          github-token: ${{ secrets.GITHUB_TOKEN }}

github-token is what enables the action to comment the preview URL on the PR. Omit it (or set github-comment: false) if you don't want the comment.

Common patterns

Production deploy on push to main

yaml
on:
  push:
    branches: [main]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: amondnet/vercel-action@v42
        with:
          vercel-token: ${{ secrets.VERCEL_TOKEN }}
          vercel-org-id: ${{ secrets.VERCEL_ORG_ID }}
          vercel-project-id: ${{ secrets.VERCEL_PROJECT_ID }}
          vercel-args: --prod

Prebuilt — Method 1: build outside the action

Build with the Vercel CLI in a prior step, then deploy the prebuilt output. Best when you need to share the build artifact with other steps (tests, smoke checks).

yaml
- run: npm install --global vercel@latest
- run: vercel pull --yes --environment=preview --token=${{ secrets.VERCEL_TOKEN }}
  env:
    VERCEL_ORG_ID: ${{ secrets.VERCEL_ORG_ID }}
    VERCEL_PROJECT_ID: ${{ secrets.VERCEL_PROJECT_ID }}
- run: vercel build --token=${{ secrets.VERCEL_TOKEN }}
  env:
    VERCEL_ORG_ID: ${{ secrets.VERCEL_ORG_ID }}
    VERCEL_PROJECT_ID: ${{ secrets.VERCEL_PROJECT_ID }}
- uses: amondnet/vercel-action@v42
  with:
    vercel-token: ${{ secrets.VERCEL_TOKEN }}
    vercel-org-id: ${{ secrets.VERCEL_ORG_ID }}
    vercel-project-id: ${{ secrets.VERCEL_PROJECT_ID }}
    prebuilt: true

The npm install --global vercel@latest step is required on a fresh runner — the Vercel CLI is not preinstalled on GitHub-hosted runners.

Prebuilt — Method 2: let the action build

vercel-build: true runs vercel pull + vercel build inside the action and then deploys the prebuilt output. Simpler — one step total — but you can't reuse the build artifact elsewhere.

yaml
- uses: amondnet/vercel-action@v42
  with:
    vercel-token: ${{ secrets.VERCEL_TOKEN }}
    vercel-org-id: ${{ secrets.VERCEL_ORG_ID }}
    vercel-project-id: ${{ secrets.VERCEL_PROJECT_ID }}
    vercel-build: true

Alias domains with placeholders

alias-domains accepts one domain per line and substitutes {{PR_NUMBER}} and {{BRANCH}}. The domains must already exist on the Vercel project.

yaml
- uses: amondnet/vercel-action@v42
  with:
    vercel-token: ${{ secrets.VERCEL_TOKEN }}
    vercel-org-id: ${{ secrets.VERCEL_ORG_ID }}
    vercel-project-id: ${{ secrets.VERCEL_PROJECT_ID }}
    alias-domains: |
      pr-{{PR_NUMBER}}.preview.example.com
      {{BRANCH}}.preview.example.com

When alias domains are present the preview-url output is the first alias rather than the raw deployment URL.

GitHub Deployments

yaml
permissions:
  contents: read
  deployments: write

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: amondnet/vercel-action@v42
        id: vercel
        with:
          vercel-token: ${{ secrets.VERCEL_TOKEN }}
          vercel-org-id: ${{ secrets.VERCEL_ORG_ID }}
          vercel-project-id: ${{ secrets.VERCEL_PROJECT_ID }}
          github-token: ${{ secrets.GITHUB_TOKEN }}
          github-deployment: true
          # Optional override; otherwise auto-derived from vercel-args.
          # github-deployment-environment: staging
      - run: echo "Deployment id ${{ steps.vercel.outputs.deployment-id }}"

Deployment lifecycle: created in_progress before deploy, then transitioned to success (with the preview/alias URL as the environment URL) or failure. Previous deployments to the same environment are auto-deactivated.

Input reference (essentials)

InputRequiredNotes
vercel-tokenyesVercel personal token.
vercel-org-idyesRequired for Vercel CLI 17+. Works for both CLI and API modes — prefer over scope.
vercel-project-idyesRequired for Vercel CLI 17+.
vercel-argsnoAd-hoc CLI flags (e.g. --prod, --force). Mutually exclusive with experimental-api and vercel-build.
working-directorynoRun the deploy from a subdirectory.
alias-domainsnoMulti-line list; supports {{PR_NUMBER}} and {{BRANCH}}.
prebuiltnoDeploy a prebuilt .vercel/output directory. Mutually exclusive with vercel-build.
vercel-buildnoRun vercel pull + vercel build inside the action. Mutually exclusive with prebuilt and vercel-args.
vercel-output-dirnoOverride the prebuilt output path. Defaults to {working-directory}/.vercel/output when prebuilt: true.
github-tokennoNeeded for PR/commit comments.
github-commentnotrue (default), false, or a custom comment string.
github-deploymentnotrue to create a GitHub Deployment. Requires permissions: deployments: write.
github-deployment-environmentnoOverride the auto-derived environment name.

For the long tail (target, force, env, build-env, regions, archive, root-directory, auto-assign-custom-domains, custom-environment, public, with-cache, scope, vercel-version, vercel-project-name, experimental-api) read action.yml — every input there is the source of truth.

Outputs

OutputSet whenNotes
preview-urlalwaysRaw deployment URL, or the first alias domain if alias-domains is set.
preview-namealwaysDeployment name (resolved via vercel inspect).
deployment-idonly when github-deployment: trueGitHub Deployment ID — useful for follow-up status updates.

Mutual exclusions and gotchas

  • experimental-api: true cannot be combined with vercel-args. The API client uses @vercel/client, an internal Vercel package without semver guarantees — keep CLI mode (the default) unless the user explicitly wants the typed inputs (target, regions, env, build-env, public, with-cache, custom-environment, auto-assign-custom-domains).
  • prebuilt cannot be combined with vercel-build. Pick exactly one prebuilt flow.
  • vercel-build: true cannot be combined with a non-empty vercel-args. The action throws at config-parse time with a clear error — vercel-build deploys the locally produced .vercel/output via the prebuilt path, while vercel-args routes through the CLI path which would ignore that output.
  • scope is CLI-mode only. Prefer vercel-org-id so the same workflow works in both modes.
  • The auto-detected GitHub Deployment environment is mode-dependent. CLI mode: production if vercel-args contains --prod or --production, otherwise preview. Experimental API mode: derived from target (production or preview). Set github-deployment-environment explicitly for staging and similar in either mode.
  • Deprecated input aliases (still accepted, emit deprecation warnings): zeit-token → vercel-token, now-args → vercel-args, now-project-id → vercel-project-id, now-org-id → vercel-org-id. Always rewrite to the modern names.
  • The action sets VERCEL_TELEMETRY_DISABLED=1 and exports VERCEL_ORG_ID / VERCEL_PROJECT_ID from the corresponding inputs — don't redeclare them in env: for the action step itself.

References

  • action.yml — full input/output contract (source of truth).
  • README.md — additional examples and a longer changelog.
  • https://vercel.com/account/tokens — where to mint the VERCEL_TOKEN secret.

Frequently asked questions

What does the Use Vercel Action AI skill do?

Wire `amondnet/vercel-action` into a GitHub Actions workflow to deploy Vercel projects from CI. Use when the user asks to deploy to Vercel from GitHub Actions, add Vercel preview deploys, post a Vercel preview URL on a PR, configure prebuilt Vercel deployments, set up alias domains for previews, create a GitHub Deployment for Vercel, or migrate off Vercel's native GitHub integration. Covers required tokens/IDs, the CLI vs experimental-API modes, prebuilt flows, alias placeholders, deprecated `zeit-*`/`now-*` aliases, and the action's outputs (`preview-url`, `preview-name`, `deployment-id`).

Why use Use Vercel Action on TypingMind?

Because you install it once and use it with any model. Use Vercel Action is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Use Vercel Action in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/amondnet/vercel-action/tree/master/skills/use-vercel-action. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Use Vercel Action?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Use Vercel Action?

As many as you like. As long as a model supports skills, you can use Use Vercel Action with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Use Vercel Action AI skill free?

Yes. It is published on GitHub by amondnet under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇