Plasmic for AI agents
Plasmic is a visual builder for the web, letting teams create and manage pages and components without writing code. Marketing and product teams use it to ship web pages faster. 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 Plasmic.
Trigger a page action
Run a configured global action on a Plasmic page, such as submitting a form.
Check a component's configuration read-only
Look up how a component is wired before triggering an action on it.
Confirm an action before running it live
Require confirmation before triggering an action that changes published page content.
1 Plasmic actions, graded by risk.
Every Plasmic action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
0Look things up. Allowed by default.
- No read actions.
write
1Create and change things. Allow, or ask the user first.
- plasmic.global_actions_providerPlasmic Global Actions Provider
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
Plasmic in three steps.
- 01Your users connect PlasmicThey add their Plasmic api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, and writes like “plasmic Global Actions Provider” can wait for the user to approve.
- 03Any agent can actYour agent calls Plasmic through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('plasmic', { read: 'allow', write: 'ask', // global_actions_provider destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Plasmic connects.
Users add their Plasmic 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 Plasmic 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 Plasmic from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
Plasmic and Arc0, answered.
Can I use Plasmic with Claude, ChatGPT or Cursor?
Yes. Connect Plasmic 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 Plasmic actions you allow.
How do users connect Plasmic?
Users add their Plasmic 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 Plasmic actions can my agent take?
1 in total: 0 read, 1 write and 0 destructive, such as “plasmic Global Actions Provider”. Your policies decide which of them each agent may call.
Can I make my agent read-only in Plasmic?
Yes. Allow read actions and deny writes in the Plasmic policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call Plasmic too?
Yes. The same Plasmic connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Plasmic into your agent.
Your users connect Plasmic once, under your brand. Your agent gets 1 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan