Connect the Lucid MCP server to Unblocked to give AI agents live diagram context

System diagrams, sequence flows, and whiteboard sessions often hold the clearest picture of how a feature was meant to work, but they live in Lucid where coding agents can’t see them. Lucid’s MCP server lets AI agents search and read your Lucid documents, and Unblocked connects to it once so every agent can consult a diagram without its own Lucid setup.

This is a third-party MCP server: Lucid 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 diagrams as they are now.

Wiring Lucid’s MCP server straight into an agent gives it the shapes and text on a page. Unblocked’s Context Engine reasons across those results and your other sources — following a box on an architecture diagram to the service that implements it, the pull request that reshaped it, and the ticket that linked the diagram as its design — and returns one cited answer.

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

What you get

• Diagrams agents can read — Agents search for Lucid documents and read their text and structure, so a flowchart or architecture diagram becomes context instead of a picture nobody opens.
• One connection for every agent — Connect Lucid to Unblocked once. Claude Code, Cursor, Codex, and GitHub Copilot reach it through the Unblocked MCP server, with no per-agent Lucid authorization.
• View-only access to your documents — Lucid’s server can also create and edit documents. Unblocked offers only its read-only tools, and you choose which of them it may call from Settings → Data Sources.
• Design intent matched to the code — The Context Engine reconciles Lucid diagrams with your code, pull requests, and conversations, so the design an engineer drew and the system that shipped can be compared — with citations.

Prerequisites

• An Unblocked account (free to start).
• A Lucid account with access to the documents you want to query; on Team and Enterprise plans, a Lucid admin must first enable MCP access. FedRAMP accounts aren’t supported.
• An MCP-compatible coding agent (Claude Code, Cursor, Devin, etc.) to use the context in your editor.

Step 1: Connect Lucid to Unblocked

1. In Unblocked, go to Settings → Data Sources → Connect another data source, and under Documentation select Lucid.
2. The Lucid MCP server URL is filled in for you. Choose OAuth, click Connect, and sign in to Lucid 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

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

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

• context_researchResearches a question across your connected sources and calls the Lucid tools you selected when your diagrams or their 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 Lucid data:

• Find the Lucidchart diagram for our order pipeline and tell me whether the queue it shows still exists in the code.
• What does the onboarding flow diagram linked in this Jira epic describe, and which pull requests implemented it?
• Are there unresolved comments on the auth sequence diagram, and were they discussed in Slack?
• Which services appear on our network architecture diagram but have no recent commits?