Tooly
Tooly

Vercel

Deployment and hosting tools powered by Vercel API

The @tooly/vercel package provides AI-ready tools for Vercel deployment and hosting platform. Manage projects, deployments, domains, and teams with AI assistance.

Installation

npm install @tooly/vercel

Quick Start

import { generateText } from 'ai'
import { openai } from '@ai-sdk/openai'
import { createAITools } from '@tooly/vercel'

const tools = createAITools(process.env.VERCEL_BEARER_TOKEN!)

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'List my Vercel projects and show me their latest deployments',
    },
  ],
  tools,
})

console.log(result.text)

Setup

1. Get Your Vercel Bearer Token

  1. Go to Vercel Account Settings
  2. Create a new access token
  3. Copy the token for use in your application

2. Environment Variables

Store your bearer token securely:

VERCEL_BEARER_TOKEN=your_vercel_bearer_token_here

3. Initialize the Tools

import { createAITools } from '@tooly/vercel'

const tools = createAITools(process.env.VERCEL_BEARER_TOKEN!)

Available Tools

The Vercel package provides the following AI tools:

createProject

Creates a new project in Vercel.

Parameters:

  • name (string, required): Project name
  • framework (string, optional): Framework preset (e.g., "nextjs", "vite", "create-react-app")
  • gitRepository (object, optional): Git repository configuration
    • type (string): Git provider ("github", "gitlab", "bitbucket")
    • repo (string): Repository name (e.g., "username/repo-name")
  • publicSource (boolean, optional): Whether the project source code is public
  • rootDirectory (string, optional): Root directory relative to repository root
  • teamId (string, optional): Team ID to create the project under
  • slug (string, optional): Team slug to create the project under

Example:

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'Create a new Next.js project called "my-app" connected to my GitHub repo',
    },
  ],
  tools,
})

getProject

Gets details of a specific Vercel project by ID or name.

Parameters:

  • idOrName (string, required): Project ID or name
  • teamId (string, optional): Team ID that owns the project
  • slug (string, optional): Team slug that owns the project

Example:

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'Show me details for my project called "portfolio"',
    },
  ],
  tools,
})

listProjects

Lists all Vercel projects.

Parameters:

  • teamId (string, optional): Filter by team ID
  • slug (string, optional): Filter by team slug
  • limit (number, optional): Maximum number of projects to return (1-100)
  • since (number, optional): Get projects created after this timestamp
  • until (number, optional): Get projects created before this timestamp

Example:

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'List all my Vercel projects',
    },
  ],
  tools,
})

updateProject

Updates an existing Vercel project.

Parameters:

  • idOrName (string, required): Project ID or name to update
  • name (string, optional): New project name
  • framework (string, optional): New framework preset
  • publicSource (boolean, optional): Whether the project source code is public
  • rootDirectory (string, optional): New root directory
  • teamId (string, optional): Team ID that owns the project
  • slug (string, optional): Team slug that owns the project

Example:

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'Update my project to use Vite framework and make the source public',
    },
  ],
  tools,
})

deleteProject

Deletes a Vercel project.

Parameters:

  • idOrName (string, required): Project ID or name to delete
  • teamId (string, optional): Team ID that owns the project
  • slug (string, optional): Team slug that owns the project

Example:

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'Delete my old test project',
    },
  ],
  tools,
})

getDeployment

Gets details of a specific deployment by ID or URL.

Parameters:

  • idOrUrl (string, required): Deployment ID or URL
  • teamId (string, optional): Team ID that owns the deployment
  • slug (string, optional): Team slug that owns the deployment

Example:

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'Show me details for deployment dpl_abc123',
    },
  ],
  tools,
})

listDeployments

Lists deployments for a project or team.

Parameters:

  • projectId (string, optional): Filter by project ID
  • teamId (string, optional): Team ID to list deployments for
  • slug (string, optional): Team slug to list deployments for
  • limit (number, optional): Maximum number of deployments to return (1-100)
  • since (number, optional): Get deployments created after this timestamp
  • until (number, optional): Get deployments created before this timestamp
  • state (string, optional): Filter by deployment state (e.g., "BUILDING", "READY", "ERROR")
  • target (string, optional): Filter by deployment target ("staging", "production")

Example:

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'Show me the latest production deployments for my project',
    },
  ],
  tools,
})

listProjectDomains

Lists domains for a specific project.

Parameters:

  • idOrName (string, required): Project ID or name
  • teamId (string, optional): Team ID that owns the project
  • slug (string, optional): Team slug that owns the project

Example:

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'Show me all domains configured for my portfolio project',
    },
  ],
  tools,
})

getTeam

Gets details of a specific team.

Parameters:

  • teamId (string, required): Team ID

Example:

const result = await generateText({
  model: openai('gpt-4.1-nano'),
  messages: [
    {
      role: 'user',
      content: 'Show me details for my team',
    },
  ],
  tools,
})

Integration Examples

OpenAI SDK

import OpenAI from 'openai'
import { createOpenAIFunctions } from '@tooly/vercel'

const openai = new OpenAI()
const { functions, callFunction } = createOpenAIFunctions(process.env.VERCEL_BEARER_TOKEN!)

const response = await openai.chat.completions.create({
  model: 'gpt-4',
  messages: [
    {
      role: 'user',
      content: 'Create a new project for my portfolio website',
    },
  ],
  functions,
})

if (response.choices[0].message.function_call) {
  const result = await callFunction(response.choices[0].message.function_call)
  console.log(result)
}

Anthropic SDK

import { createAnthropicTools } from '@tooly/vercel'

const { tools, callTool } = createAnthropicTools(process.env.VERCEL_BEARER_TOKEN!)

// Use with Anthropic Claude API
// Implementation depends on your Anthropic integration

Error Handling

All tools include built-in error handling and will throw descriptive errors for common issues:

  • Invalid bearer token
  • Missing required parameters
  • API rate limits
  • Network connectivity issues
  • Invalid project or deployment IDs
try {
  const result = await generateText({
    model: openai('gpt-4.1-nano'),
    messages: [{ role: 'user', content: 'List my projects' }],
    tools,
  })
} catch (error) {
  console.error('Vercel operation failed:', error.message)
}

Best Practices

  1. Secure Token Storage: Store your Vercel bearer token securely using environment variables
  2. Rate Limiting: Be mindful of Vercel API rate limits when making frequent requests
  3. Error Handling: Always implement proper error handling for production applications
  4. Team Management: Use team IDs and slugs to organize projects within team accounts
  5. Deployment Monitoring: Regularly check deployment status and logs for issues

Rate Limits

Vercel API has rate limits that vary by plan:

  • Hobby: 100 requests per hour
  • Pro: 1,000 requests per hour
  • Enterprise: Custom limits

The package automatically handles rate limiting errors and provides meaningful error messages.

Support

For issues with the Vercel package:

  1. Check the Vercel API documentation
  2. Verify your bearer token permissions
  3. Review the error messages for specific guidance
  4. Open an issue on the Tooly GitHub repository

For Vercel platform issues, contact Vercel Support.

PayPal

Payment processing tools powered by PayPal API

Custom

Base abstractions and utilities for building AI tool packages