Set up the MCP server in Claude Code
Connect Claude Code to BugMojo over MCP so it can read a bug's replay context, logs, and repro steps and start fixing.
BugMojo exposes a hosted MCP server so AI coding agents can read a bug the way a developer would: its session replay, console logs, network requests, and repro steps. This guide connects Claude Code to that server in one command, then authorizes and verifies it.
What you'll need
Before you start, make sure you have the following:
- Claude Code installed and working in your terminal.
- A BugMojo account you can sign in to in a browser.
- Access to at least one bug. The agent inherits your permissions, so you only see bugs you already have access to.
Add the BugMojo MCP server
Open a terminal in the project you want Claude Code to work in.
Run the add command below. It registers a server named bugmojo that talks to the hosted endpoint over HTTP.
Start (or restart) Claude Code so it picks up the new server.
claude mcp add --transport http bugmojo https://www.bugmojo.com/api/mcpAuthorize and verify the connection
In Claude Code, ask the agent to look at a bug. For example, paste a BugMojo bug link or ask it to list your open bugs.
The first BugMojo tool call opens your browser to authorize. Sign in and approve the connection so the agent can act on your behalf.
Run /mcp inside Claude Code and confirm bugmojo is listed as connected. You should now see BugMojo tools available.
Common questions
Do I need to install a separate package?
No. BugMojo runs a hosted HTTP MCP endpoint. The claude mcp add command points Claude Code straight at it. There is nothing to install locally.
How does Claude Code authenticate to BugMojo?
The first time Claude Code calls a tool, BugMojo opens a browser to authorize the connection against your account. Sign in and approve, and the agent inherits your access to bugs.
The server does not show up after I add it. What now?
Restart Claude Code so it reloads its MCP config, then run /mcp again. If bugmojo is still missing, re-run the add command and check the transport is http.
Related articles
- How AI agents work with BugMojoUnderstand the full loop, from a captured bug to an agent-proposed fix that a human verifies and closes.
- Set up the BugMojo MCP server in CursorConnect Cursor to BugMojo over MCP so its AI can read a bug's replay, logs, and repro steps.
- MCP tool referenceKnow exactly which tools a connected AI agent can call over MCP to read a BugMojo bug and act on it.
- Agent access, tokens, and the security modelUnderstand how an AI agent's access is scoped, what it inherits from you, and the guardrails that keep it in bounds.