Build and debug Bird messaging inside GitHub Copilot and VS Code.

Connect GitHub Copilot to Bird so your agent can configure sending domains, wire transactional email, SMS, WhatsApp, OTP verification, and Voice into your codebase, and inspect bounces or active calls without leaving VS Code.

Get StartedContact Sales

Drag an icon onto the dots. Use arrow keys to move a focused icon. Press Escape to cancel a drag.

From prompt to delivered.

Send messages and check delivery from your code, terminal, or AI assistant.

YouVoice

Can you check which calls are active right now and tell me which are connected and which are still ringing?

Bird MCP4 connected actions
  1. 1Read the workspaceworkspace_get
  2. 2Find calls in progressvoice_legs_list
  3. 3Inspect the connected callvoice_legs_get
  4. 4Inspect the ringing callvoice_legs_get
Explore MCP docs

Connect GitHub Copilot and VS Code to your Bird workspace in four steps.

Install the Bird skills for GitHub Copilot, add the hosted MCP server in VS Code, and sign in through your browser so your agent can inspect senders, configure channels, and test delivery.

Install in VS Code Read the setup guide
Terminal · Plugin
copilot plugin marketplace add messagebird/bird-ai
copilot plugin install bird@bird-ai
.vscode/mcp.json
{
  "servers": {
    "bird": {
      "type": "http",
      "url": "https://mcp.bird.com"
    }
  }
}

  1. Install the Bird skills

    Run the plugin commands in your terminal to install the Bird CLI and email audit skills for GitHub Copilot. These skills give your agent workflow guidance for configuring channels, auditing sending domains, and writing integration code.

    Without the plugin, run npx skills add messagebird/bird-ai to install the same two skills, then add the MCP server in the next step.

  2. Register the MCP server in VS Code

    Select Install in VS Code above, or save the server configuration block in your project workspace so teammates share the same connection setup.

    You can also add the server entry to your personal VS Code user configuration to keep Bird available across projects.

  3. Sign in to your workspace

    Open the MCP server list from the VS Code command palette and start the Bird server. Complete the browser sign-in to grant GitHub Copilot agent mode access to your workspace.

  4. Verify your active workspace

    Ask GitHub Copilot to confirm your signed-in account and active Bird workspace before making changes. Once your workspace details return, your agent is ready to inspect domains, configure senders, and test messages.

Run messaging and voice tasks from GitHub Copilot chat.

Ask your agent to diagnose email bounces, configure SMS or WhatsApp senders, wire OTP verification into your app, or inspect live Voice calls using your real workspace state.

Understand a delivery issue
Check my recent email failures. Group them by reason, inspect the affected sending domain, and tell me what needs fixing before making changes.
Prepare a new channel
Help me set up SMS for order updates. Check my workspace and available senders, explain what is missing, then prepare a test for a number I provide.
See what is happening now
List the Voice calls that are active right now. Separate connected calls from calls that are still ringing, and show their IDs.

How authentication and workspace permissions work.

What to know about browser sign-in, delegated workspace access, and application credentials when using Bird with GitHub Copilot in VS Code.

Does connecting MCP give GitHub Copilot unrestricted access to my workspace?

No. GitHub Copilot operates strictly within the Bird role and workspace permissions you hold and approve during browser sign-in. You can revoke the session from your Bird account settings whenever you need to.

Do I need an API key to use Bird in VS Code?

No. Connecting GitHub Copilot to the hosted Bird MCP server uses interactive browser sign-in. When your agent writes application code that sends email, SMS, WhatsApp, or Voice traffic from your own backend, that deployed service uses its own Bird access key.

What if the Bird tools appear in VS Code but requests fail?

Verify that you completed browser sign-in after adding the server configuration in VS Code. Open the MCP server list from the command palette, restart or reauthorize the Bird connection, and ask your agent to confirm your active workspace before running channel or domain tasks.

Extend your Bird workflow with hosted MCP tools, reusable agent skills, and other AI editors.

MCP server

Inspect how hosted HTTP transport, browser OAuth sign-in, and scoped workspace permissions protect live Bird operations.

Agent skills

Install reusable skills that guide your agent through domain audits, sender configuration, and multi-channel API code.

Other AI tools

Connect the same Bird workspace to Cursor, Claude Code, Windsurf, OpenAI Codex, and other MCP-compatible clients.

Start building with Bird in minutes.

Get StartedContact Sales