Bubble for AI agents
Bubble is a visual, no-code platform for building web apps, where records called Things and drag-built workflows stand in for hand-written backend code. 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 Bubble.
Create a new data record
Add a new Thing to a Bubble app's database, such as a customer record or order entry.
Trigger an app workflow
Kick off a Bubble workflow through its API endpoint, like starting a welcome email sequence.
Confirm a record before deleting it
Fetch a Thing by its ID to confirm it's the right one before it gets deleted.
15 Bubble actions, graded by risk.
Every Bubble 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
14Create and change things. Allow, or ask the user first.
- bubble.file_uploadUpload File
- bubble.file_downloadDownload File
- bubble.oauth_authorizeOAuth Authorize
- bubble.file_temp_createCreate Temp File
- bubble.data_create_thingCreate Thing
- bubble.o_auth_access_tokenGet OAuth Access Token
- bubble.o_auth_register_appBubble OAuth Register App
- bubble.data_get_thing_by_idGet Thing By ID
- bubble.workflow_trigger_getTrigger Bubble Workflow via GET
- bubble.meta_get_swagger_jsonGet Bubble API Swagger JSON
- bubble.workflow_trigger_postTrigger Bubble Workflow via POST
- bubble.data_patch_thing_by_idPatch Thing By ID
- bubble.data_bulk_create_thingsBulk Create Things
- bubble.data_put_replace_thing_by_idReplace Thing by ID
destructive
1Delete, cancel or archive. Ask first, or deny outright.
- bubble.data_delete_thing_by_idDelete Thing By ID
Bubble in three steps.
- 01Your users connect BubbleThey add their Bubble api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “upload File” can wait for the user, and “delete Thing By ID” can be denied outright.
- 03Any agent can actYour agent calls Bubble through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('bubble', { read: 'allow', write: 'ask', // file_upload destructive: 'deny', // data_delete_thing_by_id }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Bubble connects.
Users add their Bubble 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 Bubble 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 Bubble from any agent.
Bubble and Arc0, answered.
Can I use Bubble with Claude, ChatGPT or Cursor?
Yes. Connect Bubble 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 Bubble actions you allow.
How do users connect Bubble?
Users add their Bubble 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 Bubble actions can my agent take?
15 in total: 0 read, 14 write and 1 destructive, such as “upload File”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Bubble?
Yes. Actions like “delete Thing By ID” 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 Bubble too?
Yes. The same Bubble connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Bubble into your agent.
Your users connect Bubble once, under your brand. Your agent gets 15 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan