V0 for AI agents
v0 is Vercel's AI tool for generating production-ready web code from a prompt, used by developers who want a working component or page to start from. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in V0.
Generate code from a prompt
Ask v0 to build a component or page and get back working code.
Find an existing chat or project
Look up a previous v0 chat or project as a read-only search before continuing it.
Confirm before deploying a project
Get approval before deploying generated code, since a deployment makes it publicly reachable.
44 V0 actions, graded by risk.
Every V0 action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
24Look things up. Allowed by default.
- v0.get_chatGet Chat
- v0.get_hookGet Hook
- v0.get_userGet User
- v0.find_chatsFind Chats
- v0.list_hooksList Hooks
- v0.get_messageGet Chat Message
- v0.get_projectGet Project by ID
- v0.find_projectsFind Projects
- v0.list_messagesList Messages
- v0.get_rate_limitsGet Rate Limits
- v0.get_user_scopesGet User Scopes
- v0.get_chat_projectGet Chat Project
- v0.get_user_billingGet User Billing
- v0.get_user_plan_v0Get User Plan
- v0.list_deploymentsList Deployments
- v0.list_chat_versionsList Chat Versions
write
15Create and change things. Allow, or ask the user first.
- v0.update_chatUpdate Chat
- v0.update_hookUpdate V0 Webhook
- v0.create_hook_v0Create Webhook
- v0.create_projectCreate V0 Project
- v0.update_projectUpdate V0 Project
- v0.update_chat_versionUpdate Chat Version Files
- v0.create_vercel_projectCreate Vercel Project
- v0.create_project_env_varsCreate Project Environment Variables
- v0.update_project_env_varsUpdate Project Environment Variables
- v0.fork_chatFork Chat
- v0.init_v0_chatInitialize Chat
- v0.favorite_chatFavorite Chat
- v0.deploy_projectDeploy Project
- v0.chat_completionsV0 Chat Completions
- v0.assign_chat_to_projectAssign Chat To Project
destructive
5Delete, cancel or archive. Ask first, or deny outright.
- v0.delete_chatDelete Chat
- v0.delete_hookDelete Hook
- v0.delete_deploymentDelete Deployment
- v0.delete_v0_projectDelete V0 Project
- v0.delete_project_env_vars_v0Delete Project Environment Variables
V0 in three steps.
- 01Your users connect V0They add their V0 api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Webhook” can wait for the user, and “delete Chat” can be denied outright.
- 03Any agent can actYour agent calls V0 through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('v0', { read: 'allow', write: 'ask', // create_hook_v0 destructive: 'deny', // delete_chat }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How V0 connects.
Users add their V0 api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same V0 connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use V0 from any agent.
V0 and Arc0, answered.
Can I use V0 with Claude, ChatGPT or Cursor?
Yes. Connect V0 to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the V0 actions you allow.
How do users connect V0?
Users add their V0 api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which V0 actions can my agent take?
44 in total: 24 read, 15 write and 5 destructive, such as “create Webhook”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in V0?
Yes. Actions like “delete Chat” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call V0 too?
Yes. The same V0 connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug V0 into your agent.
Your users connect V0 once, under your brand. Your agent gets 44 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan