Pumble for AI agents
Pumble is a team communication platform offering channels, direct messages, and scheduled messages for workspace collaboration. Teams use it as a chat hub for daily coordination. 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 Pumble.
Send a message to a channel
Post an update to a team channel so everyone sees the latest status.
Search messages read-only
Search past messages for a keyword to find a decision made earlier.
Delete a message with approval
Remove a sent message only after the author confirms it should come down.
9 Pumble actions, graded by risk.
Every Pumble action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
6Look things up. Allowed by default.
- pumble.list_usersList Users
- pumble.get_messageGet Message
- pumble.list_channelsList Channels
- pumble.search_messagesSearch Messages
- pumble.get_current_userGet Current User
- pumble.list_channel_messagesList Channel Messages
write
2Create and change things. Allow, or ask the user first.
- pumble.send_messageSend Message
- pumble.edit_messageEdit Message
destructive
1Delete, cancel or archive. Ask first, or deny outright.
- pumble.delete_messageDelete Message
Pumble in three steps.
- 01Your users connect PumbleThey add their Pumble api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “send Message” can wait for the user, and “delete Message” can be denied outright.
- 03Any agent can actYour agent calls Pumble through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('pumble', { read: 'allow', write: 'ask', // send_message destructive: 'deny', // delete_message }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Pumble connects.
Users add their Pumble 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 Pumble 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 Pumble from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
Pumble and Arc0, answered.
Can I use Pumble with Claude, ChatGPT or Cursor?
Yes. Connect Pumble 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 Pumble actions you allow.
How do users connect Pumble?
Users add their Pumble 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 Pumble actions can my agent take?
9 in total: 6 read, 2 write and 1 destructive, such as “send Message”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Pumble?
Yes. Actions like “delete Message” 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 Pumble too?
Yes. The same Pumble connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Pumble into your agent.
Your users connect Pumble once, under your brand. Your agent gets 9 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan