Use this guide to connect BugHerd to other AI tools that support MCP (Model Context Protocol), so you can access projects, tasks, and comments directly from your AI workflows.
This guide covers AI tools we don't have a dedicated support doc for. The exact steps will vary depending on your AI tool, but the overall process is the same.
Prerequisites
Before connecting, ensure:
You have a BugHerd account and You have access to at least one BugHerd organization
You are logged into your AI tool of choice
Your AI tool supports MCP (Model Context Protocol) connections – check your tool's documentation if you're unsure
BugHerd MCP can be used with MCP-compatible clients that support either:
Remote Streamable HTTP MCP servers, or
Local stdio MCP servers via a bridge, such as
mcp-remote
BugHerd’s remote MCP server URL is:
https://mcp.bugherd.com/mcp
BugHerd uses OAuth authentication, so the client must be able to open the BugHerd authorization page and store the resulting access token.
Option 1: Direct remote MCP setup
Use this option if your client supports remote HTTP or Streamable HTTP MCP servers.
Add BugHerd using the following details:
Setting | Value |
Name | BugHerd |
URL | |
Transport | HTTP / Streamable HTTP |
After adding the server, authenticate when prompted:
Open the BugHerd authorization page
Log in if needed
Select the organization you want the client to access
Click Allow
Return to your MCP client
Option 2: Local setup using npx mcp-remote
Use this option if your MCP client only supports local stdio servers.
mcp-remote acts as a local bridge between stdio-only MCP clients and remote MCP servers, including OAuth-enabled servers.
Additional prerequisites
You’ll need:
Node.js installed
npxavailable in your terminalAn MCP-compatible client that supports local command-based MCP servers
Check Node and npx:
node -v
npx -v
Generic MCP configuration
Add the following to your client’s MCP configuration:
{
"mcpServers": {
"BugHerd": {
"command": "npx",
"args": [
"-y",
"mcp-remote@latest",
"https://mcp.bugherd.com/mcp"
],
"env": {}
}
}
}Save the configuration and restart your AI tool.
Test the connection
Try:
“List my BugHerd projects”
If your projects are returned, the connection is working.
What you can do with BugHerd's MCP
For a full list of what you can do with BugHerd's MCP, check out Getting Started with the BugHerd MCP server.
Need help? Contact [email protected] or start a chat from the app.