Skip to main content
The Bolna MCP server connects your AI assistant directly to your Bolna account over the Model Context Protocol. Once connected, you can list and inspect agents, pull call transcripts, check your wallet balance, create or update agents, and place real outbound calls — all from a chat window, no dashboard tab required. It’s hosted at mcp.bolna.ai — there’s nothing to install or run locally. Point your client at the URL with your API key and you’re connected.

Quickstart

Connect Claude, Cursor, Windsurf, Codex, Zed, or any other MCP client in under a minute

​
Who it’s for

The same 11 tools serve two different ways of working — nothing about the server itself is coding-specific:

Anyone on the team

Chat-only clients like Claude Desktop are enough. Check a wallet balance, look up an agent, pull a transcript, or fix a prompt — all in plain English, no dashboard needed.

Developers using a coding agent

In Claude Code, Cursor, or Windsurf, the same tools ground the assistant in your real agent IDs and call data while it writes code against the REST API — see the dashboard example.

​
What you can do

“List my agents”, “what’s my wallet balance?”, “show me the numbers on my account” — quick lookups without opening the dashboard.
“Get the last 5 executions for agent X and show me the transcript for any that failed” — the assistant chains the lookups for you.
“Create a Hindi lead-qualification agent using Sarvam and GPT-4o mini”, “update agent X’s welcome message” — describe the change, the assistant makes the API call.

​
How it works

Your client sends your Bolna API key as a Bearer token to https://mcp.bolna.ai/api/mcp. The server checks the key, calls the matching Bolna REST API endpoint, and hands back trimmed JSON your assistant can reason about.
Diagram showing an MCP client connecting via Bearer token to mcp.bolna.ai, which calls api.bolna.ai for agents, calls, phone numbers, batches, and account

How the Bolna MCP server connects your AI assistant to your account

No key is stored server-side — it travels with each request. The server is open source; see the GitHub repo for the implementation.

​
Confirm before it acts

update_agent, delete_agent, and start_outbound_call are flagged as destructive in their tool definitions. Most clients ask you to confirm before running them — delete_agent is irreversible, and start_outbound_call places a live call and spends account balance.

​
Reference

Tool List

Every tool the server exposes, grouped by agents, calls, phone numbers, and account

Prompt Cheatsheet

Copy-paste prompts for common tasks — works in any client, chat-only included

​
For coding agents

Example: Monitoring Dashboard

Requires Claude Code, Cursor, Windsurf, or another agent that can write and run files — use the MCP server to explore your data, then have it build a small dashboard around it

Bolna Skills

A code-native alternative — teaches the same operations to Claude Code, Cursor, and Codex directly, no MCP connection needed