Deploy to Vercel
A deployment skill, not just advice: at use time it really deploys the user's project to Vercel.
- What
- A deployment skill, not just advice: at use time it really deploys the user's project to Vercel.
- Cost
- Free
- Needs
- Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else.
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — A deployment skill, not just advice: at use time it really deploys the user's project to Vercel. It first checks the project's state (git remote, .vercel/project.json or .vercel/repo.json link files, whether the Vercel CLI is installed and authenticated, the available teams), then deploys by git push when the project is linked with a git remote, or with vercel deploy directly — preview by default, production only when the user explicitly asks. Where the CLI cannot be installed or authenticated (claude.ai or Codex sandboxes), it falls back to bundled shell scripts that package the project (excluding node_modules, .git and .env), upload it to a claimable-deploy endpoint and return a preview URL plus a claim URL. Installing the skill itself deploys nothing and needs no account; all of the deployment power is at use time, on the user's own Vercel account. By @vercel-labs, listed here with credit to its creator. MIT-licensed (announced in the upstream README; no LICENSE file). Skill Harbor never reviews the code, review it yourself before use.
Version:
Install
Copy the install package below, then paste it into Muse📌 Pinned version: commit 063bee9Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites — Nothing to install the skill itself — no account, no key, no payment, and installing it deploys nothing. The warning belongs to use time, stated up front: this skill really deploys. When later used, it deploys my project to Vercel through the authenticated Vercel CLI on my own account (installing the CLI with npm install -g vercel and running vercel login if needed), choosing the team with me and asking before any git push; or, in a sandbox where the CLI cannot authenticate, through its bundled scripts, which package my project (excluding node_modules, .git and .env) and upload it to a claimable deploy endpoint. Preview deployments by default; production only if I explicitly ask. Install "Deploy to Vercel" for me. It is the deployment skill for Vercel: at use time it gathers the project's state (git remote, Vercel link files, CLI authentication, teams), picks the best deployment path — git push for a linked project with a remote, vercel deploy for a linked project without one, link-then-deploy when the CLI is authenticated but the project is not linked, install-and-login when the CLI is missing — and always deploys as a preview unless I explicitly ask for production, returning the deployment URL. In sandboxes where the CLI cannot authenticate, it falls back to its bundled deploy scripts, which package and upload the project to a claimable deploy endpoint and return a preview URL and a claim URL. Its folder holds the main SKILL.md (frontmatter name deploy-to-vercel), the two fallback shell scripts under resources/ (deploy.sh for the claude.ai sandbox, deploy-codex.sh for Codex), and an Archive.zip whose listing shows copies of the same SKILL.md and scripts plus macOS metadata files. Repository: https://github.com/vercel-labs/agent-skills (the skill is the whole skills/deploy-to-vercel/ folder at the pinned commit: SKILL.md plus 3 companion files, 4 files in total) Pinned version: install the code at commit 063bee94c3f4df8453406c830b0a7df0f2860278 exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/vercel-labs/agent-skills into a temporary folder, check out commit 063bee94c3f4df8453406c830b0a7df0f2860278, and copy the skills/deploy-to-vercel/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-22: [FINGERPRINTS] - Archive.zip: SHA-256 37c0d035b3d2c6256105316a2492c1487f3688efa21f030a8a623a9f7f0ee02f - SKILL.md: SHA-256 cfcc3dd479ab2e0ae721ddf39b8af84d977321487672f1487c8d6855f576927b - resources/deploy-codex.sh: SHA-256 18b411e2bfe07778bca11d7f0e1ec9c95e5764011dd2e3838f51066728646855 - resources/deploy.sh: SHA-256 418da6ff48c582790799d850a88fd83ca00f2ee00502665caf8dbb06d4dab8b9 If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full, and tell me in plain words: what the skill does when used, what it connects to, whether it reads or sends any personal information or credentials, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the files: - The skill folder holds 4 files: SKILL.md (frontmatter name deploy-to-vercel), resources/deploy.sh and resources/deploy-codex.sh (the no-auth fallback deploy scripts), and Archive.zip, an archive whose listing shows copies of the same SKILL.md and scripts plus macOS metadata files. The scripts are copied as-is and are never run during installation. - Nothing is sent or deployed anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit. No project of mine is read, packaged or uploaded during installation. - At use time the skill deploys for real on my own Vercel account: it checks git and link state and CLI authentication, asks me to choose the team when several exist and asks before any git push, deploys preview by default (production only on my explicit request), or — in a sandbox without CLI auth — uploads my packaged project (excluding node_modules, .git and .env) to a claimable deploy endpoint and returns preview and claim URLs. - Do not install the Vercel CLI, do not log in, do not link, push or deploy anything, and do not run either deploy script during this install: using the skill is a separate step that only starts when I ask for a deployment later. 3. When I say "yes, install", install it as a skill: copy the whole deploy-to-vercel folder into the agent's skills directory, in a folder named exactly "deploy-to-vercel", keeping its 4 files in their original relative layout (SKILL.md at its root). 4. Verify with no network calls: frontmatter valid (name: deploy-to-vercel), the 4 files in place with matching fingerprints. Do not run anything from the folder and do not use the skill to verify. 5. Report what was installed, where, and what I still need to do myself (for example: that using it will deploy my project to Vercel for real (preview by default, production only on my explicit request) through my authenticated CLI or the sandbox fallback scripts, and that using it is a separate step that only starts when I ask for a deployment). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Saved to your recent installs. Find it anytime on /connect.
🔒 Security, provenance and what this touches
- It can download and install software packages.
What the scan found — excerpts from the install prompt
- installs packages — Prerequisites — Nothing to install the skill itself — no account, no key, no payment, and installing it deploy · Prérequis — Rien pour installer le skill lui-même — aucun compte, aucune clé, aucun paiement, et l'installer n
The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.