Connect the Miro MCP server to Unblocked to give AI agents live board and diagram context

Architecture sketches, sequence diagrams, and planning boards often live in Miro, far from the code they describe. An agent working in the editor can’t see the board where the team drew the service boundaries or left review comments on a flow, so it builds without the design behind the task. Miro’s MCP server exposes boards to AI agents, and Unblocked connects to it once so every agent can read them without its own Miro setup.

This is a third-party MCP server: Miro hosts and maintains it, and Unblocked connects to it as a verified MCP server. Nothing is ingested or indexed — Unblocked calls the read-only tools you select at the moment a question needs them, so answers reflect your boards as they are now, within the board permissions of the account that authorized the connection.

Wiring Miro’s MCP server straight into an agent gives it board items and text. Unblocked’s Context Engine reasons across those results and your other sources — following a frame in an architecture diagram to the pull request that implemented it, the ticket that scoped it, and the Slack thread where the design changed — and returns one cited answer.

Once connected, any MCP-compatible coding agent — Claude Code, Cursor, Devin, GitHub Copilot, Codex — reaches Miro context through a single Unblocked MCP server connection.

What you get

• Live boards, diagrams, and docs — Agents read frames, diagrams, docs, tables, and comments as they are on the board right now, through Miro’s own MCP server rather than an exported snapshot.
• One connection for every agent — Connect Miro to Unblocked once. Claude Code, Cursor, Codex, and GitHub Copilot reach it through the Unblocked MCP server, with no per-agent Miro configuration.
• Read-only, board-safe control — Only read-only tools are offered, so agents can’t create items, edit docs, or resolve comments on your boards. Choose which tools Unblocked may call from Settings → Data Sources.
• Context Engine reasoning — Miro results are reconciled with your code, pull requests, tickets, and conversations, so a diagram on a board leads to the implementation and the decisions behind it — with citations.

Prerequisites

• An Unblocked account (free to start).
• A Miro account with access to the boards you want to query and permission to authorize an OAuth app; on Enterprise plans, an admin must first enable the Miro MCP server for the organization.
• An MCP-compatible coding agent (Claude Code, Cursor, Devin, etc.) to use the context in your editor.

Step 1: Connect Miro to Unblocked

1. In Unblocked, go to Settings → Data Sources → Connect another data source, and under Documentation select Miro.
2. The Miro MCP server URL is filled in for you. Choose OAuth, click Connect, and sign in to Miro to authorize Unblocked.
3. Once Unblocked discovers the server’s tools, select the read-only tools it may use and click Save Settings.

Step 2: Install the Unblocked MCP server

On macOS or Linux, one command auto-detects and configures the supported agents on your machine:

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

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

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

https://getunblocked.com/api/mcpsse

Miro hosts and maintains this MCP server. Unblocked calls only the read-only tools you select, at the moment a question needs them:

• board_search_boardsSearches and lists the Miro boards the connected account can access.
• context_exploreSurveys a board’s high-level items — frames, docs, tables, and diagrams — so a question can target the right part of the board.
• context_getReturns text context from a whole board or a single item. Calls to this tool use Miro AI credits.
• board_list_itemsLists the items on a board, with cursor-based pagination and optional filters.
• layout_readReads existing board items and returns them as Miro’s layout DSL text.
• comment_list_commentsLists the comments on a board or on a specific item.

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

Unblocked acts as the MCP context layer for your Miro data, searching boards and diagrams alongside your other connected sources and returning cited results through these tools:

• context_researchResearches a question across your connected sources and calls the Miro tools you selected when boards, diagrams, or board comments are relevant, returning a synthesized answer with citations.
• context_get_urlsFetches the full content behind URLs your agent encounters across connected sources — Jira, Linear, Notion, Confluence, Slack, GitHub, and more.

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

• Does the payments service match the architecture diagram on our platform Miro board, and which pull requests diverged from it?
• What open comments are on the checkout flow board, and is there a ticket tracking each one?
• Summarize the sequence diagram for the auth handshake and point me to the code that implements it.
• Which services on the migration planning board have shipped, and what did the team say in Slack about the ones that haven’t?