Deploy to Vercel
Deploys a project to Vercel, as a preview unless told otherwise, through the Vercel CLI or, without an account, a bundled script that uploads the folder and returns a preview link and a claim link. Use when you want a shareable deployment link fast.
vercel-deploy- Author
- OpenAI
- Source
- openai/skills/skills/.curated/vercel-deploy @ 49f948f
- License
- MIT · Copyright (c) 2026 Vercel
About this skill
What it does
The skill deploys a project to Vercel and returns the link. It deploys a preview unless you explicitly ask for production. If the Vercel CLI is installed (command -v vercel), the agent runs vercel deploy [path] -y with a ten-minute timeout, adding --prod only for production.
When the CLI is missing or has no credentials, the agent runs scripts/deploy.sh, which needs no Vercel account. The script copies the project to a temporary folder without node_modules, .git, .env and .env.*, detects the framework from package.json (over forty, from Next.js and Astro to Hono and Express), packs a tarball and posts it to codex-deploy-skills.vercel.sh. It polls the preview URL for up to five minutes and prints JSON with previewUrl and claimUrl; the claim URL moves the deployment into your own Vercel account. A lone HTML file without package.json is renamed to index.html.
The agent returns the links and is told not to fetch the deployed URL to check it. Besides SKILL.md and the script, the folder holds agents/openai.yaml, icons and Vercel's MIT licence.
When to use it, and when not
Use it for "deploy this and give me a link": a prototype, a demo, a preview to share.
On a private project, note that the fallback uploads everything except those four patterns, so a credentials.json or a key file would go too. SKILL.md says nothing about environment variables, domains or Git integration; set those up in Vercel yourself.
How to run it
The install command below puts the whole folder in place. The fallback needs a Bash shell with tar and curl. In a Codex sandbox only the deploy command, never the CLI check, is rerun with escalated permissions (sandbox_permissions=require_escalated) when network calls are blocked. The agent picks the skill up by its description; to call it yourself:
- Claude Code:
/vercel-deploy, or ask for the skill by name. - Codex:
$vercel-deploy. - Cursor: type
/in the Agent chat and pickvercel-deploy.
Example: /vercel-deploy Deploy this folder as a preview and give me the link.
OpenAI has marked openai/skills deprecated; our copy is pinned to a commit there, and as of November 2026 openai/plugins has no newer version of this skill.
How it differs from similar skills
Deploy to Vercel, Vercel Labs has a word-for-word identical description but a longer flow: it checks the Git remote and the project link first and steers you towards deploys by git push. The agent cannot tell the two apart, so install one. Vercel CLI with tokens works from an access token instead of a browser login and also manages environment variables, domains and logs.
Deploy to Netlify always needs a Netlify login, links the folder to a site and documents netlify.toml, redirects and functions; this skill is shorter and can deploy without an account. Deploy to Render deploys from Git, including services with databases and background workers. Web frameworks on Cloudflare Workers targets Cloudflare Workers and has no deploy command of its own: it leaves the commands to the Wrangler and cf skills. CI/CD pipeline setup builds a preview deploy per pull request into the pipeline instead of deploying from the chat.
Install
Run one command — it saves SKILL.md and the files it uses into your agent's skills folder. Claude Code and Codex pick up a new skill without a restart; if it does not show up, restart the agent.
curl -fsSL --create-dirs -o ~/.claude/skills/vercel-deploy/SKILL.md https://anymodel.org/en/skills/openai/vercel-deploy/SKILL.md -o ~/.claude/skills/vercel-deploy/LICENSE.txt https://anymodel.org/en/skills/openai/vercel-deploy/LICENSE.txt -o ~/.claude/skills/vercel-deploy/agents/openai.yaml https://anymodel.org/en/skills/openai/vercel-deploy/agents/openai.yaml -o ~/.claude/skills/vercel-deploy/assets/vercel-small.svg https://anymodel.org/en/skills/openai/vercel-deploy/assets/vercel-small.svg -o ~/.claude/skills/vercel-deploy/assets/vercel.png https://anymodel.org/en/skills/openai/vercel-deploy/assets/vercel.png -o ~/.claude/skills/vercel-deploy/scripts/deploy.sh https://anymodel.org/en/skills/openai/vercel-deploy/scripts/deploy.sh
curl -fsSL --create-dirs -o ~/.agents/skills/vercel-deploy/SKILL.md https://anymodel.org/en/skills/openai/vercel-deploy/SKILL.md -o ~/.agents/skills/vercel-deploy/LICENSE.txt https://anymodel.org/en/skills/openai/vercel-deploy/LICENSE.txt -o ~/.agents/skills/vercel-deploy/agents/openai.yaml https://anymodel.org/en/skills/openai/vercel-deploy/agents/openai.yaml -o ~/.agents/skills/vercel-deploy/assets/vercel-small.svg https://anymodel.org/en/skills/openai/vercel-deploy/assets/vercel-small.svg -o ~/.agents/skills/vercel-deploy/assets/vercel.png https://anymodel.org/en/skills/openai/vercel-deploy/assets/vercel.png -o ~/.agents/skills/vercel-deploy/scripts/deploy.sh https://anymodel.org/en/skills/openai/vercel-deploy/scripts/deploy.sh
foreach ($f in 'SKILL.md','LICENSE.txt','agents/openai.yaml','assets/vercel-small.svg','assets/vercel.png','scripts/deploy.sh') { $p = Join-Path "$HOME\.claude\skills\vercel-deploy" $f; New-Item -ItemType Directory -Force (Split-Path $p) | Out-Null; Invoke-WebRequest "https://anymodel.org/en/skills/openai/vercel-deploy/$f" -UseBasicParsing -OutFile $p }foreach ($f in 'SKILL.md','LICENSE.txt','agents/openai.yaml','assets/vercel-small.svg','assets/vercel.png','scripts/deploy.sh') { $p = Join-Path "$HOME\.agents\skills\vercel-deploy" $f; New-Item -ItemType Directory -Force (Split-Path $p) | Out-Null; Invoke-WebRequest "https://anymodel.org/en/skills/openai/vercel-deploy/$f" -UseBasicParsing -OutFile $p }Any other agent: save the file into its skills folder, or give it the link to the file: https://anymodel.org/en/skills/openai/vercel-deploy/SKILL.md
SKILL.md
Vercel Deploy
Deploy any project to Vercel instantly. Always deploy as preview (not production) unless the user explicitly asks for production.
Prerequisites
- Check whether the Vercel CLI is installed without escalated permissions (for example,
command -v vercel). - Only escalate the actual deploy command if sandboxing blocks the deployment network calls (
sandbox_permissions=require_escalated). - The deployment might take a few minutes. Use appropriate timeout values.
Quick Start
- Check whether the Vercel CLI is installed (no escalation for this check):
command -v vercel
- If
vercelis installed, run this (with a 10 minute timeout):
vercel deploy [path] -y
Important: Use a 10 minute (600000ms) timeout for the deploy command since builds can take a while.
- If
vercelis not installed, or if the CLI fails with "No existing credentials found", use the fallback method below.
Fallback (No Auth)
If CLI fails with auth error, use the deploy script:
skill_dir="<path-to-skill>"
# Deploy current directory
bash "$skill_dir/scripts/deploy.sh"
# Deploy specific project
bash "$skill_dir/scripts/deploy.sh" /path/to/project
# Deploy existing tarball
bash "$skill_dir/scripts/deploy.sh" /path/to/project.tgz
The script handles framework detection, packaging, and deployment. It waits for the build to complete and returns JSON with previewUrl and claimUrl.
Tell the user: "Your deployment is ready at [previewUrl]. Claim it at [claimUrl] to manage your deployment."
Production Deploys
Only if user explicitly asks:
vercel deploy [path] --prod -y
Output
Show the user the deployment URL. For fallback deployments, also show the claim URL.
Do not curl or fetch the deployed URL to verify it works. Just return the link.
Troubleshooting
Escalated Network Access
If deployment fails due to network issues (timeouts, DNS errors, connection resets), rerun the actual deploy command with escalated permissions (use sandbox_permissions=require_escalated). Do not escalate the command -v vercel installation check. The deploy requires escalated network access when sandbox networking blocks outbound requests.
Example guidance to the user:
The deploy needs escalated network access to deploy to Vercel. I can rerun the command with escalated permissions—want me to proceed?
Connect through AnyModel API
This skill runs on whatever model your agent uses. Point Claude Code, Codex or Cursor at AnyModel to run it on Claude, GPT, Gemini and more with one key — pay per token, no subscription.
AnyModel