developer-toolsoauth2

Vercel

Frontend deploys, preview URLs, analytics.

Verdict

Vercel via MCP is the integration that lets the model see your deploys, preview URLs, and project settings. Most useful for frontend teams shipping multiple times a day where "what's the status of the deploy" is a question that comes up. What we notice: read operations are the day-to-day win — list recent deploys, show the build log for a failure, fetch the preview URL for a branch, find which env vars are set. The model can answer "why did this deploy fail" usefully if it can pull the build log. Write operations (trigger redeploys, change env vars) work but rarely need the AI in the loop — those are typically run from the CLI. Best for: frontend teams running multiple PRs through Vercel's preview-deploy flow; debugging failed deploys via build log analysis; project status summaries for non-engineering stakeholders; finding the right preview URL for a stakeholder review. Avoid for: complex deployment orchestration that involves multiple tools beyond Vercel; workflows where the Vercel CLI or dashboard already covers the team's needs; orgs where Vercel access is restricted by role and the MCP would breach the boundary. Practical frame: free tier covers personal use; team plans unlock the API. Token cost is small — most Vercel objects are compact. The compounding win comes from "I don't have to context-switch into the Vercel dashboard to find the preview URL."

Common use cases

  • Deploy from chat after code review
  • Add secrets before production push
  • Check domain availability during naming discussion
  • Verify cache artifacts exist before build
  • Generate auth tokens for CI pipeline

Integration

Vendor
Vercel
Category
developer-tools
Auth
OAUTH2
Tools
35
Composio slug
vercel

Tools

  • Add Environment Variable

    Tool to add an environment variable to a vercel project. use after confirming the project exists and you need to configure secrets or configuration values across environments before deployment. example: "add api key=secret to production".

  • Check Cache Artifact Exists

    Tool to check if a cache artifact exists by its hash. use when verifying whether a cache artifact is already stored before upload.

  • Check Domain Availability

    Tool to check if a domain is available for registration. use when you need to verify domain availability before purchase.

  • Check Domain Price

    Tool to check the price for a domain before purchase. use when evaluating cost and availability prior to domain registration.

  • Create Auth Token

    Tool to create a new authentication token. use when you need to programmatically generate a new token after validating permissions. example: "create auth token named my-token"

  • Create Edge Config

    Tool to create a new edge config for a vercel project. use when you need to define edge caching settings before deploying your project across a specified repository.

  • Create Edge Config Token

    Tool to create a new token for a specific edge config. use when you need a read-only access token after provisioning your edge config.

  • Create new deployment

    Tool to create a new deployment. use when you need to deploy files or a git commit to a vercel project.

  • Delete Auth Token
    destructive

    Tool to delete an authentication token. use when you need to revoke a token programmatically after confirming its validity. example: "delete auth token with id abc123"

  • Delete Deployment
    destructive

    Tool to delete a specific deployment by its unique id. use after confirming the deployment identifier to clean up unused or failed deployments.

  • Delete Edge Config Tokens
    destructive

    Tool to delete tokens associated with a specific edge config. use when you need to revoke one or more access tokens from an existing edge config by its id.

  • Delete Environment Variable
    destructive

    Tool to delete a specific environment variable from a project. use after verifying the correct variable id to remove it.

  • Delete Vercel Project
    destructive

    Tool to delete a specific project by its id or name. use after confirming the correct project id or name to permanently remove it.

  • Deploy Edge Function

    Deploy edge functions to vercel. use when you need to deploy serverless functions that run on the edge runtime with fast cold starts and global distribution.

  • Get Auth Token Metadata

    Tool to retrieve metadata for an authentication token. use when you need to inspect details of a specific token for auditing or debugging.

  • Get deployment details

    Tool to retrieve detailed information about a specific deployment. use after triggering a deployment and you need to inspect its status and configuration. example: "get details for deployment dpl 123abc".

  • Get Deployment Events

    Tool to retrieve events related to a specific deployment. use when monitoring or debugging deployment history or streaming real-time events. example: "get events for deployment dpl xxx since 1540095775941."

  • Get Deployment Logs

    Tool to retrieve logs for a specific vercel deployment. use when monitoring deployment execution, debugging issues, or analyzing deployment performance. example: "get logs for deployment dpl xxx since 1540095775941."

  • Get Domain Transfer Info

    Tool to get information required to transfer a domain to vercel. use when you need to check transfer availability or current status before initiating a transfer.

  • Get Edge Config

    Tool to retrieve details of a specific edge config. use when you need to inspect edge config metadata by id before updating or deleting.

  • Get Edge Config Item

    Tool to retrieve a specific item within an edge config. use after obtaining the edge config id and when you need to inspect or validate a particular configuration item by its key.

  • Get Edge Config Token

    Tool to retrieve details of a specific token associated with an edge config. use when you need metadata for an existing edge config token.

  • Get Vercel Project

    Tool to retrieve information about a vercel project by id or name. use when you need project metadata after obtaining its identifier.

  • List All Deployments

    Tool to list all deployments. use after authenticating to retrieve deployments under your user or team context. example: "list deployments for project qmx...".

  • List All Teams

    Tool to list all teams accessible to the authenticated user. use after authentication to retrieve their ids and slugs.

  • List Auth Tokens

    Tool to list authentication tokens. use when you need to retrieve all tokens for the current user or an optional team.

  • List Deployment Checks

    Tool to retrieve a list of checks for a specific deployment. use after a deployment to inspect check statuses and results.

  • List Edge Config Items

    Tool to retrieve a list of items within a specific edge config. use when you need to inspect or manage all key–value items after creating an edge config.

  • List Edge Configs

    Tool to list all edge configs. use when you need to enumerate all edge config definitions for your account or team.

  • List Edge Config Tokens

    Tool to retrieve a list of tokens for a specific edge config. use after obtaining an edge config's id to list its tokens.

  • List Environment Variables

    Tool to list environment variables for a specific project. use when you need to inspect or page through the environment settings before deployment.

  • List Vercel Aliases

    Tool to list all aliases for the authenticated user or team. use when you need to retrieve aliases filtered by various criteria.

  • Update Edge Config

    Tool to update an existing edge config. use when you need to rename the slug of an edge config for reorganization or migration purposes. only the slug field may be updated; ensure you confirm the new slug before using.

  • Update Edge Config Items

    Tool to update items within a specific edge config. use when you need to batch modify, add, or remove key-value pairs in an existing edge config.

  • Update Vercel Project

    Tool to update an existing project. use when you need to modify a vercel project’s configuration before deployment. use after confirming the project id or name.

Setup

Setup guide

  1. 11. In Switchy, open your workspace settings and click 'Add Integration' under the MCP section. 2. Select Vercel from the developer tools category. 3. Click 'Connect with Vercel' to start OAuth — you'll authorize Switchy to read and write projects, deployments, domains, and environment variables on your Vercel account. 4. Choose which Vercel team or personal account to connect if you belong to multiple. 5. Grant the requested scopes (projects, deployments, env vars, domains, edge configs) and confirm. 6. Back in Switchy, you'll see 'Vercel connected' with a green checkmark. 7. Open any Space, type '@Vercel check domain availability for myapp.com' and send — if you get a response listing availability and price, the connection works. 8. To deploy, @mention Vercel with a project name and Git reference, like '@Vercel create new deployment for my-nextjs-app from main branch'.

What teammates see: by default, memories from Vercel are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.

Works well with

Top models

Compatibility data appears once enough Spaces have used this MCP together with a given model.

How Switchy teams use it

Not enough Spaces yet to publish anonymised usage stats (we require ≥ 50 Spaces per week).

Starter prompts

Deploy Latest Commit

@Vercel create new deployment for my-nextjs-app from the main branch to production
Open in a Space →

Add Environment Variable

@Vercel add environment variable STRIPE_SECRET_KEY with value sk_live_abc123 to my-nextjs-app for production and preview
Open in a Space →

Check Domain Price

@Vercel check domain price for myapp.com and tell me if it's available
Open in a Space →

Verify Cache Artifact

@Vercel check if cache artifact exists with hash a1b2c3d4e5f6 in my-nextjs-app
Open in a Space →

Create Auth Token

@Vercel create auth token named github-actions-deploy with read and write permissions
Open in a Space →

Example outputs

Illustrative - representative of the model's voice and quality, not literal recordings.

Prompt

@vercel list all deployments for my-app from the last 7 days and show their status

Output

Here are the recent deployments for my-app: 1. Production (main branch) - deployed 2 hours ago - READY URL: my-app.vercel.app Commit: feat: add user dashboard (abc123) 2. Preview (feature/auth) - deployed yesterday - READY URL: my-app-git-feature-auth.vercel.app Commit: wip: oauth integration (def456) 3. Production (main branch) - deployed 3 days ago - READY URL: my-app.vercel.app Commit: fix: mobile layout (ghi789) All deployments completed successfully. The most recent production build is serving traffic.

Notes

This example demonstrates read-only project inspection. The MCP requires a valid API token with at least read scope for the target project. Response shape varies by your actual deployment history — this illustrative output shows typical metadata (branch, commit, URL, status) but won't match your account state.

Prompt

@vercel add environment variable DATABASE_URL to my-app for production and preview environments

Output

Environment variable added successfully: Variable: DATABASE_URL Project: my-app Environments: Production, Preview Type: Encrypted (secret) The variable is now available to all new deployments in these environments. Existing deployments will need to be redeployed to pick up this change. Would you like me to trigger a new deployment now?

Notes

This example shows a write operation that modifies project configuration. The MCP requires write permissions on the project. Important: adding env vars does NOT automatically redeploy — you must trigger a new deployment for the variable to take effect. Secrets are encrypted at rest but visible to anyone with project access.

Prompt

@vercel check the last 5 production deployments for my-app and summarize any failures or performance regressions

Output

Analysis of the last 5 production deployments for my-app: ✓ 4 successful deployments (builds completed in 45-52 seconds) ✗ 1 failed deployment (2 days ago): - Build error: TypeScript compilation failed in src/utils/api.ts - Commit: refactor: update API client (xyz789) - Fixed in next deployment 3 hours later Performance trend: Build times stable around 48 seconds average. No significant regressions detected. The failed deployment was quickly resolved and didn't impact production traffic due to automatic rollback.

Notes

This example pairs the MCP's deployment data with AI reasoning to surface patterns. The MCP fetches raw deployment metadata (status, duration, errors), and the AI synthesizes it into actionable insights. Useful for post-mortems or release health checks. Requires read access and works best with projects that have consistent deployment history.

Use-case deep-dives

Preview deploys for design handoff

When Vercel MCP speeds up frontend iteration cycles

A 5-person product team ships React prototypes twice a week for stakeholder review. The designer updates Figma, the frontend dev pushes a branch, and the PM needs a live URL in Slack within 10 minutes. The Vercel MCP handles this loop from Switchy: create deployment from the branch, check status, grab the preview URL, and drop it in the thread. No context-switching to the Vercel dashboard. The win is real if your team deploys more than 3 previews per week and the PM or designer isn't comfortable with CLI tools. If you're a solo dev or your deploy cadence is monthly, the overhead of configuring 35 tools outweighs the convenience. For fast-moving small teams where non-engineers need deploy visibility, this MCP closes the handoff gap.

Environment variable rotation after breach

How this MCP handles emergency secret updates at scale

Your startup's API key leaks in a public repo on Friday afternoon. You have 12 Next.js projects on Vercel, each with 4 environments (prod, preview, dev, staging). Manually rotating 48 environment variables through the dashboard takes 90 minutes and invites typos. The Vercel MCP's Add Environment Variable tool lets you script the rotation from Switchy: loop through projects, confirm each exists, push the new secret to all environments, trigger redeployments. The entire operation runs in under 10 minutes with an audit trail in the chat. This scenario justifies the MCP if you manage more than 5 Vercel projects or rotate secrets quarterly. Below that threshold, the dashboard is faster than learning the tool names.

Domain purchase for client project kickoff

When domain tooling in Switchy beats the registrar dashboard

A 3-person agency onboards a new client and needs to register a domain, check pricing, and provision DNS before the kickoff call in 2 hours. The Vercel MCP's Check Domain Availability and Check Domain Price tools let the account manager validate options and quote costs without leaving Switchy or pinging the dev lead. Once the client approves, the same MCP creates the project and wires up the domain in one session. The trade-off: Vercel's domain pricing runs 15-20% higher than Namecheap, so this workflow makes sense only if your team values consolidated billing and same-day DNS propagation over cost. For agencies running 10+ client projects per year on Vercel, the time savings justify the premium.

Frequently asked

What does the Vercel MCP let me do in Switchy?

It lets your AI agents deploy projects, manage environment variables, configure edge caching, check domain availability, and create auth tokens — all without leaving the chat. You can spin up a deployment from a Git commit, add secrets to a project, or verify a domain is free to register, just by asking. It's Vercel's platform API exposed as conversational tools.

Do I need a Vercel API key to connect this MCP?

Yes. The MCP uses API key authentication, so you'll need to generate a token from your Vercel account settings. Anyone on your team with a Vercel account can connect it, but the token inherits the permissions of whoever creates it — so if you want agents to deploy to production, use a token from someone with deploy access.

Can the Vercel MCP roll back a deployment or delete a project?

No. The 35 tools focus on creating and configuring resources — deployments, environment variables, edge configs, domains, tokens. There's no rollback or delete capability exposed through this MCP. If you need to undo something, you'll still use the Vercel dashboard or CLI directly.

Why use this instead of just running vercel deploy in my terminal?

Because your AI agent can now deploy on your behalf mid-conversation. If you're debugging in Switchy and the agent suggests a config change, it can add the environment variable and trigger the deployment without you context-switching to a terminal. It's faster for iterative workflows where the AI is already making decisions about your infrastructure.

Who on my team should connect the Vercel MCP?

Whoever owns your deployment pipeline. The API key they use determines what the agent can touch — if you connect it with a personal token, agents can only act on projects that person can access. For shared team projects, consider a service account token with scoped permissions, so you're not tying deployments to one engineer's credentials.

Data last verified 7 hours ago.Sources aggregated hourly to weekly. See docs/architecture/model-directory.md.