Connect the Omni MCP server to Unblocked to give AI agents live BI and semantic model context

When a dashboard number looks wrong, the answer is split between Omni and the codebase. The model and topic definitions live in Omni, while the events, pipelines, and schema changes that feed them live in code, so an agent asked about a metric can see only half the picture. Omni’s MCP server exposes its models, topics, queries, and saved content to AI agents, and Unblocked connects to it once so every agent can ask Omni without its own setup.

This is a third-party MCP server: Omni 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 query results reflect your Omni models and data as they are now, under the permissions of the user who authorized the connection.

Wiring Omni’s MCP server straight into an agent gives it query results. Unblocked’s Context Engine reasons across those results and your other sources — following a shifted dashboard metric to the topic it is built on, the pull request that changed the underlying table, and the Slack thread where someone first flagged the drop — and returns one cited answer.

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

What you get

• Queries through your semantic model — Agents pick an Omni model and topic, then query it, so answers use the joins and definitions your data team already maintains instead of hand-written SQL against raw tables.
• Saved dashboards and queries in reach — Agents search your Omni dashboards and apps and run the saved queries inside them, so a question can start from the chart the team already trusts.
• Read-only, tool-level control — Unblocked offers only Omni’s read-only tools — nothing that edits dashboards, apps, or drafts. Omni admins also control which capabilities the server allows, and you choose which tools Unblocked may call from Settings → Data Sources.
• Context Engine reasoning — Omni results are reconciled with your code, pull requests, tickets, and conversations, so an unexpected number leads to the change that caused it — with citations.

Prerequisites

• An Unblocked account (free to start).
• An Omni organization where an admin has enabled AI, Omni Agent, and the MCP server in AI Hub, and an Omni user who can authorize through OAuth.
• An MCP-compatible coding agent (Claude Code, Cursor, Devin, etc.) to use the context in your editor.

Step 1: Connect Omni to Unblocked

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

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

• pickModelLists the Omni models available to the authorized user, with their IDs.
• pickTopicLists the topics in a model and picks the most relevant one for a question.
• getDataRuns a query against a chosen model and topic and returns the results.
• searchContentSearches your Omni dashboards and apps.
• runSavedQueryRuns a saved query from an Omni document by its ID and returns the results.
• searchOmniDocsSearches Omni’s documentation and returns an answer with citations to specific pages.

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

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

• context_researchResearches a question across your connected sources and calls the Omni tools you selected when models, dashboards, or query results 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 Omni data:

• Why did weekly signups drop on the growth dashboard, and did a recent pull request change the events table it reads from?
• Which Omni topic defines net revenue, and does it match the logic in our billing service?
• Run the saved churn query from the retention dashboard and tell me whether anyone has opened a ticket about the spike.
• Before I rename this column, which Omni dashboards would break, and who discussed the change in Slack?