How to connect Vercel to OpenClaw
Connect Vercel to OpenClaw with ClawLink: paste your Vercel key once and 131 tools work from chat. No MCP server to run, nothing stored on your machine.


Manage projects, deployments, domains, and environment variables. Once connected, OpenClaw can read and act on Vercel from chat — pairing, token refresh, and tool wiring handled for you.
The usual route to Vercel access for OpenClaw is an MCP server you configure and keep running, plus your own OAuth app or API keys. ClawLink gives OpenClaw a more practical Vercel setup: install one ClawHub skill, connect Vercel in the browser, and OpenClaw can call real Vercel actions from any chat surface with no auth, token refresh, or tool wiring to build yourself.
Copy this prompt into OpenClaw, or open the Vercel skill on ClawHub.
Before installing anything, inspect the ClawHub skill metadata and setup requirements.
If the skill asks you to install a third-party package or CLI, verify its source, maintainer, and package contents before running the install command.
Install the skill "Vercel" (hith3sh/vercel-deployments) from ClawHub only after those checks pass.
Skill page: https://clawhub.ai/hith3sh/vercel-deployments
Keep the work scoped to this skill only.
After install, help me finish setup from verified skill metadata.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.Setup
It takes three steps to connect OpenClaw to Vercel.
1Install the plugin
Paste the setup prompt into OpenClaw, or install from the terminal and ask OpenClaw to pair:
openclaw plugins install clawhub:clawlink-plugin- 2
Connect Vercel
Paste your API key in the dashboard.
- 3
Use it from chat
Ask OpenClaw: "What can you do with Vercel?"
Install by command
The setup prompt above does all of this in one paste. By hand, it is one install command plus a browser approval:
openclaw plugins install clawhub:clawlink-pluginThen ask OpenClaw to set up ClawLink. It starts browser pairing and prints an approval link — open it, approve the device, return to the chat, and say done. Finally, connect Vercel in the ClawLink dashboard — paste your Vercel API key once.
Verify the connection by asking OpenClaw:
List my projects with vercel_get_projects, then show me the recent deployments of the one I pick with vercel_get_deployments and flag anything that failed.
Using a different agent?
The OpenClaw plugin is one client of ClawLink's MCP server. Claude Code, Cursor, Codex, or any agent that can run a shell command pairs with the same ClawLink account through the CLI:
npx -y @useclawlink/cli loginlogin opens the same browser approval and stores a credential locally. Once Vercel is connected in the dashboard, that agent calls the same 131 Vercel tools over MCP. Full setup for MCP clients and shell agents: connect apps to any AI agent.
Vercel MCP for OpenClaw
Looking for a Vercel MCP server for OpenClaw? ClawLink connects Vercel to OpenClaw and exposes 131 Vercel tools your agent can call over MCP, with hosted auth and nothing to run or maintain yourself. Using Hermes instead? The Hermes Vercel integration works the same way.
Vercel authenticates with an API key, so you do paste one here: the difference is what happens afterwards. The setup most guides teach has you create an API key under AI Gateway, install a CLI, and stand up a sandbox or a server, which is the route for deploying OpenClaw on Vercel rather than connecting Vercel to OpenClaw. Vercel's own no-key story is OIDC tokens, which let a deployment authenticate to Vercel's AI endpoints without rotating a secret, but that is Vercel talking to Vercel; it does not hand an agent your account's projects and deployments. Here the key goes into a hosted setup page once, OpenClaw installs one ClawHub skill, and the 131 Vercel tools below are callable from chat: no CLI, no sandbox, no environment variable file, and the key lives server-side where rotating it means changing one thing.
Is it safe to connect Vercel to OpenClaw?
The AI Overview answers "is it safe to connect Vercel to an AI agent" with the Context.ai breach of April 2026: an OAuth token for a Vercel-connected service was compromised, the incident ran for roughly eight weeks, and the attackers demanded a $2M ransom. The security press covered it as a caution about where OAuth tokens live. That is the right lens for this page, and the hosted setup is the answer to it:
- The credential is not on the machine that could be breached. In the Context.ai shape of incident, the leaked token sat in the victim's own systems. Here the Vercel token exists in one hosted place, the agent never handles it, and nothing on your laptop or your OpenClaw host can leak it.
- Revocation is one action. An API key pasted into env files has to be hunted down and replaced in every place it was copied; a hosted connection is revoked from the dashboard, which stops the agent's access immediately.
- The remaining risk is behavioral. The agent acts with your account's access, so treat prompts and any third-party content as untrusted input, start with reads, and require a confirmation step before writes like deployments and environment variables, which are the actions that change live systems.
The incident did not make hosted connections immune, it made them the point: tokens are safest where they are fewest, managed in one place, and revocable in one click.
What the OpenClaw Vercel integration can do
131 Vercel tools are ready for OpenClaw once the account is connected. The 30 below are the ones people reach for most; your agent can call all 131.
30 of 131 Vercel tools for OpenClaw
| Tool | What it does |
|---|---|
Add environment variable vercel_add_environment_variable | Add an environment variable to a Vercel project |
Add project domain vercel_add_project_domain | Attach a domain to a Vercel project |
Create new deployment vercel_create_new_deployment | Create a new deployment in Vercel |
Create project2 vercel_create_project2 | Create a new Vercel project |
Filter project envs vercel_filter_project_envs | List environment variables for a Vercel project |
Get auth user vercel_get_auth_user | Get the authenticated Vercel user |
Get deployment vercel_get_deployment | Get details for a specific deployment |
Get deployments vercel_get_deployments | List Vercel deployments |
Get project domains vercel_get_project_domains | List domains attached to a Vercel project |
Get project2 vercel_get_project2 | Get details for a specific Vercel project |
Get projects vercel_get_projects | List Vercel projects |
Get teams vercel_get_teams | List teams accessible to the connected Vercel account |
List domains vercel_list_domains | List domains in the connected Vercel account |
Update project2 vercel_update_project2 | Update project settings for a Vercel project |
Check cache artifact exists vercel_check_cache_artifact_exists | Check if a cache artifact exists by its hash |
Check domain availability vercel_check_domain_availability | Check if a domain is available for registration |
Check domain price2 vercel_check_domain_price2 | Check pricing for a domain including purchase, renewal, and transfer costs |
Download artifact vercel_download_artifact | Download a cache artifact from Vercel's Remote Cache by its hash |
Get active attack status vercel_get_active_attack_status | Read active attack data from Vercel Firewall for a specific project |
Get alias vercel_get_alias | Retrieve information about a Vercel alias by ID or alias name |
Get artifact info vercel_get_artifact_info | Query information about artifacts by their hashes |
Get auth token vercel_get_auth_token | Retrieve metadata for an authentication token |
Get bulk availability vercel_get_bulk_availability | Check availability for multiple domains at once |
Get cache status vercel_get_cache_status | Get the status of Remote Caching for the principal |
Get certs vercel_get_certs | Retrieve SSL/TLS certificates for the authenticated user or team |
Get configurations vercel_get_configurations | Get configurations for the authenticated user or team |
Get contact info schema vercel_get_contact_info_schema | Retrieve the contact information schema for a domain's top-level domain (TLD) |
Get deployment events2 vercel_get_deployment_events2 | Get deployment events for a specific Vercel deployment by ID or URL |
Get deployment file contents vercel_get_deployment_file_contents | Retrieve the contents of a specific file from a Vercel deployment |
Get deployment logs2 vercel_get_deployment_logs2 | Retrieve runtime logs for a specific Vercel deployment by project and deployment ID |
Try it: find the Vercel tool you need
Browse the 30 Vercel tools
Click any tool to see exactly what OpenClaw can do and copy a ready-to-use prompt.
Example prompts
List my projects with vercel_get_projects, then show me the recent deployments of the one I pick with vercel_get_deployments and flag anything that failed.
For project <project id>, show me the existing environment variables with vercel_filter_project_envs, then add the one I give you with vercel_add_environment_variable for preview environments only, and confirm the arguments before sending.
Get deployment <deployment id> with vercel_get_deployment, pull its logs with vercel_get_deployment_logs2, and tell me whether the last build succeeded and what the errors were.
List the domains on project <project id> with vercel_get_project_domains and tell me which ones are verified and which are misconfigured.
ClawLink vs. building it yourself
The alternative to ClawLink is usually manual API key setup plus your own token handling, permission troubleshooting, and tool plumbing for OpenClaw. That is fine if you want to build and maintain the integration yourself. Most teams just want Vercel working from chat.
| Manual | ClawLink | |
|---|---|---|
| Credential handling | Collect, validate, store, and rotate the Vercel API key yourself, then make sure every tool call uses the right account. | Users complete the hosted ClawLink setup once and the connected Vercel account becomes available to the agent without you building credential management. |
| Ongoing maintenance | You own refresh logic, permission debugging, environment config, and every provider-specific edge case for Vercel. | ClawLink handles the repetitive integration plumbing so your team can focus on the workflow instead of the infrastructure. |
| Agent usability | You still need to expose the right Vercel actions to the runtime in a format your agent can reliably use. | 131 tools for Vercel are already exposed through ClawLink, so the agent can read and act from chat immediately. |
ClawLink vs. Composio
Composio also exposes Vercel to AI agents. It is developer infrastructure: Python and TypeScript SDKs, an MCP server, and a catalog past 1,000 apps, aimed at teams shipping agent products. ClawLink is built for OpenClaw users instead. You install the plugin once, connect Vercel in the browser, and the 131 tools above work from chat. There is no SDK and no config file, and the Vercel key you paste at setup is stored server-side rather than kept in your environment. Choosing between them? Read the full Composio alternatives comparison.
Troubleshooting
OpenClaw installed the Vercel skill but can't call the tools
The ClawHub skill teaches OpenClaw about Vercel, but the calls run through the ClawLink plugin and your connected account. Make sure Vercel is connected in the dashboard, then start a fresh chat so OpenClaw reloads the tool catalog. If OpenClaw runs as a persistent gateway, restart it so the new tools register.
Connection succeeds but no tools appear
Reconnect Vercel from the dashboard, then start a fresh chat if the runtime still has the old tool catalog loaded.
"Tool schema not loaded yet" error when calling Vercel tools
Vercel tool schemas load on demand the first time a tool runs and are cached after that, so this error usually clears on its own: wait a few seconds and retry the same request. If every Vercel call keeps failing with it in a fresh chat, reconnect from the dashboard, and contact support if it still persists — that pattern points to a configuration problem on our side, not something you can fix by reconnecting again.
Vercel returns 403 or "permission denied" on one action while others work
Two usual causes. The connected account may not have access to the specific workspace, inbox, store, or project in the request — check that first. If access looks right, the agent may have sent a placeholder value (like "YOUR_ID" or an example id from documentation) instead of a real one: ask it to run a list or search tool first, then retry the action with a real id from those results. Most failures at this stage are one of these two, not ClawLink bugs.
Vercel returns 401 "invalid token" although the account shows as connected
Search results for "invalid token" on this provider answer about tokens you would actually see: an AI_GATEWAY_API_KEY in a deploy, a CRON_SECRET, a CLI token checked with whoami. None of those exist in a hosted connection, because the only token involved is the one you pasted at setup, held server-side. If every Vercel call fails with an invalid-token error, the stored key was regenerated or deleted in Vercel afterwards, and re-pasting the current token from the dashboard is the fix. If only some calls fail, the token is valid and the account simply lacks access to that team or project. What is not the cause is OpenClaw's gateway device token, which is what most "AUTH_TOKEN_MISMATCH" search answers actually discuss.
Ask the agent to diagnose it:
Call vercel_get_auth_user to check which Vercel account the connection acts as, then quote the exact error from the failed call. If it is an invalid-token error I will re-paste the key; do not retry before then.Vercel returns 403 or "permission denied" on one action while others work
Two usual causes, and both are about the request rather than the connection. First, the connected account may not hold the role the action needs: Vercel gates operations by team membership and role, so a member who can read deployments cannot necessarily add environment variables or transfer projects, and a team you are not part of returns a permission error rather than a not-found. Second, the agent may have sent a placeholder or guessed id (like "YOUR_ID" or a project name from documentation) instead of a real one: ask it to run vercel_get_projects first and retry with the real project id. Most 403s at this stage are one of these two, not a ClawLink bug.
Ask the agent to diagnose it:
Run vercel_get_projects and vercel_get_teams and show me the real ids. Quote the exact 403 text, then retry the failing action with a real id from that list.Vercel starts returning 429 or the agent stalls partway through an audit
Vercel rate-limits its API, and an agent walking a large team's deployments, one call per deployment, is exactly the traffic shape that meets the limit. The response is not to retry immediately but to change the shape of the work: ask for one project or one deployment rather than a full sweep, and have the agent tell you what it intends to enumerate before it starts. If a bulk audit is genuinely needed, running it in stages across a few turns is more reliable than one sweep, and reads like vercel_get_deployments exist precisely so the agent can page through results instead of fetching them one at a time.
Ask the agent to diagnose it:
Tell me how many Vercel calls you have made in the last minute and what you were enumerating. Quote any rate-limit text from the last response. Do not call Vercel again until I reply.Vercel tools are missing, or one tool name is not found
Two different failures. If OpenClaw shows no Vercel tools at all, the connection or pairing is incomplete: confirm the skill was installed, that Vercel shows as connected in the dashboard, and start a fresh chat so the tool catalog reloads. If most tools work and a single name fails, that name is wrong rather than missing, and Vercel's verb-last naming makes guesses likely (vercel_add_environment_variable, vercel_get_deployments): the error lists the closest real tools, and clawlink_list_tools --integration vercel shows the exact catalog. There is also a first-call timing case: schemas load on demand, so the opening Vercel call in a fresh session can arrive before the catalog and needs one retry.
Ask the agent to diagnose it:
List the Vercel tools you actually have access to. If there are none, run clawlink_list_integrations and say so plainly. If there are, tell me which one lists projects and use that exact name.API key setup works but results look incomplete
Double-check that the API key for Vercel has the right scopes or account access. A valid key can still be too limited for some reads or writes.
FAQ
Is there a OpenClaw Vercel integration?
Yes. ClawLink is the fastest way to connect OpenClaw to Vercel: link your Vercel account once in the browser and OpenClaw can call the Vercel API through 131 ready-made tools — no custom code or token handling.
How do I add Vercel to OpenClaw with ClawLink?
Paste the setup prompt from this page into OpenClaw. It installs the ClawLink Vercel skill from ClawHub, then you click Connect in the dashboard to authorize Vercel. OpenClaw calls the tools from the next message — no config files, and the Vercel key you paste is stored server-side instead of in your environment.
How long does it take to connect Vercel to OpenClaw?
About two minutes. Sign in, click Connect next to Vercel in the dashboard, authenticate, and OpenClaw can use it from the next chat message.
Why use ClawLink instead of wiring Vercel up myself?
The alternative to ClawLink is usually manual API key setup plus your own token handling, permission troubleshooting, and tool plumbing for OpenClaw. That is fine if you want to build and maintain the integration yourself. Most teams just want Vercel working from chat.
Does this need a Vercel API key, and where does it end up?
A Vercel API token, generated in your Vercel account settings, entered once on the hosted setup page. It carries your Vercel user's access, so the agent sees the teams, projects, and deployments that you see. Because it is stored server-side rather than in an environment file or a CLI config, it is not sitting on any machine you happen to have used, rotating it in Vercel means updating a single place, and disconnecting from the ClawLink dashboard revokes the agent's use of it immediately.
What about Vercel OIDC tokens, and how is hosted OAuth different?
Vercel's OIDC tokens are the no-secret path for a deployment inside Vercel to authenticate to Vercel's own AI Gateway: Vercel issues the token, Vercel consumes it, and there is nothing to rotate. That mechanism is not available to an agent sitting outside Vercel that wants to manage your projects, which is what this connection does. Hosted OAuth is the same idea for that job: the credential exists only server-side, the agent never sees it, and the connection is the single place it is granted and revoked. The two stories answer the same "no keys on my machine" worry from opposite ends, and this page is the agent-side one.
What can the agent do in Vercel?
The 131 tools cover the project surface: projects and their settings, deployments and their logs and file contents, environment variables, domains and their DNS records, teams and members. The writes to be deliberate about are creating deployments, adding environment variables, and changing domains, because those are live: a deployment can ship, an environment variable can break a build, and a domain change re-points traffic. Reads, starting with vercel_get_projects and vercel_get_deployments, are the safe first steps.
How is this different from running an MCP server or the vclaw CLI?
The alternative routes that search results teach are about standing up infrastructure: the vercel-openclaw deployment template, the vclaw CLI, or an MCP server you configure and keep running, each with its own token handling. Here the skill install and one dashboard connection are the entire client-side story, and Vercel sits alongside your other connected accounts with one place to revoke. Choose on how much you want to run, not on features.
OpenClaw installed the Vercel skill but can't call the tools
The ClawHub skill teaches OpenClaw about Vercel, but the calls run through the ClawLink plugin and your connected account. Make sure Vercel is connected in the dashboard, then start a fresh chat so OpenClaw reloads the tool catalog. If OpenClaw runs as a persistent gateway, restart it so the new tools register.
Related
OpenClaw GitHub integration
Manage repositories, issues, pull requests, and workflows
PagerDuty tools
Manage incidents, on-call schedules, and services
OpenClaw SafetyCulture integration
SafetyCulture (formerly iAuditor) is a workplace operations platform that enables teams to conduct digital inspections, manage audits, track issues, schedule tasks, and maintain compliance through mobile and web applications.