Bitbucket MCP integration for AI agents | Unblocked

Bitbucket integration details

Category

Source code

Indexed data

repositories, pull requests, code history

Supported products

Bitbucket Cloud · Bitbucket Data Center (Enterprise plan)

Works with

Claude Code, Cursor, Windsurf, GitHub Copilot, Codex, OpenCode, VS Code — via one MCP server connection

Content reviewed

July 20, 2026

What the Bitbucket integration adds

Bitbucket contributes two different kinds of evidence to an engineering answer: repository state and the pull-request conversations behind that state. Unblocked connects them so agents can move from an implementation detail to the review history that explains it.

Bitbucket context doesn’t live alone. Unblocked links pull requests and their discussions to the code they touch, then combines them with your docs, tickets, and conversations from every other connected source. Instead of raw search results, agents and developers get one reconciled answer with Bitbucket sources cited.

Once connected, any MCP-compatible coding agent — Claude Code, Cursor, Windsurf, GitHub Copilot, Codex — can pull this context through a single Unblocked MCP server connection.

What you get

• Repository context — Unblocked indexes your source code and keeps it up to date as you push changes. The default branch is the source of truth.
• Pull request history — Every pull request, its review comments, and merge decisions become searchable context. Agents can reference past PRs to understand why code works the way it does.
• Cloud and Data Center — Bitbucket Cloud connects via OAuth in minutes; Bitbucket Data Center is supported on the Enterprise plan with an admin install and optional service account.
• Cross-source synthesis — Bitbucket data is combined with Slack threads, Jira tickets, Confluence docs, and Sentry errors into a single knowledge graph.

How to connect Bitbucket to Unblocked

Prerequisites

• An Unblocked account (free to start).
• Bitbucket workspace administrator access to install Unblocked. Bitbucket Data Center requires the Enterprise plan.
• An MCP-compatible coding agent (Claude Code, Cursor, Windsurf, etc.) to use the context in your editor.

Step 1: Connect Bitbucket to Unblocked

1. As a Bitbucket workspace administrator, navigate to Team Settings → Data Sources in Unblocked and select Bitbucket.
2. Authorize the Unblocked app via OAuth (sign in to Bitbucket if prompted). This lets Unblocked read your workspaces and repositories but doesn’t index anything yet.
3. Select the workspace you want Unblocked to use.
4. Choose the repositories to index — Unblocked preselects repos active in the last 6 months; forks and archived repos are excluded — then click Save and Continue.

Step 2: Install the Unblocked MCP server

On macOS or Linux, one command auto-detects and configures your supported agents (Claude Code, Cursor, Windsurf, GitHub Copilot, Codex, OpenCode, VS Code):

curl -fsSL https://getunblocked.com/install-mcp.sh | bash

On Windows, or when you prefer remote MCP, add the server manually using the configuration below, or follow the manual install guide.

Claude Code

Add the remote server, then run /mcp in Claude Code, select unblocked, and log in via OAuth:

claude mcp add -s user --transport http unblocked https://getunblocked.com/api/mcpsse

Cursor

Open View: Open MCP Settings → Add Custom MCP, add the server, then click Connect to authenticate via OAuth:

{
  "mcpServers": {
    "unblocked": {
      "url": "https://getunblocked.com/api/mcpsse"
    }
  }
}

Other MCP clients

Any MCP-compatible client (Windsurf, Codex, VS Code, Claude Desktop, and others) can connect to the remote server with OAuth:

https://getunblocked.com/api/mcpsse

Connect Bitbucket to Claude Code, Cursor, and other AI agents

To give Claude Code, Cursor, GitHub Copilot, Codex, or another MCP-compatible AI coding agent access to Bitbucket context, connect Bitbucket to Unblocked and add the single Unblocked MCP server to your agent.

Unblocked acts as the MCP context layer for your Bitbucket data, searching repositories, pull requests, code history alongside your other connected sources and returning cited results through these tools:

• context_research Multi-source research across your Bitbucket repositories, pull requests, docs, and messaging. Returns a synthesized answer with citations, so agents understand not just what the code does but why it changed.
• context_get_urls Fetches the full content behind Bitbucket pull request and file URLs your agent encounters — plus Jira, Linear, Notion, Confluence, Slack, and more.

Questions to ask an AI agent about Bitbucket

Questions you (or your agent) can ask Unblocked about your Bitbucket data:

• Why was the caching layer rewritten, and which pull request introduced it?
• What decisions were made about the database migration approach?
• Which pull requests have touched the checkout service in the last 30 days?
• What did reviewers raise on the last PR to the auth repository?