---
title: Charmhub | Deploy DataHub MCP Server using Charmhub - The Open Operator Collection
description: Deploy the latest version of DataHub MCP Server as a Kubernetes Operator
  on any cloud. A Model Context Protocol server for the DataHub metadata catalog.
url: https://charmhub.io/datahub-mcp-k8s
---

# DataHub MCP Server

[Business Data](https://charmhub.io/publisher/businessdata "View all packages from Business Data")

* [Business Data](https://charmhub.io/publisher/businessdata "View all packages from Business Data")

Platform:

edge ebad6eae

```
juju deploy datahub-mcp-k8s --channel edge
```

[Learn to deploy on juju >](https://juju.is/docs/juju/manage-applications)

[Read documentation](https://github.com/canonical/canonical-data-mesh-docs)

#### Contacts

* [Submit a bug](https://github.com/canonical/datahub-mcp-k8s-operator/issues)

---

Share your thoughts on this charm with the community on discourse.

[Join the discussion](https://discourse.charmhub.io/)

#### A Model Context Protocol server for the DataHub metadata catalog.

The DataHub MCP server exposes the DataHub metadata catalog to LLM agents over the
Model Context Protocol. Agents can search datasets, walk table-level and column-level
lineage, read schemas and glossary terms, and inspect the queries that run against a
dataset, all grounded in the catalog rather than in the model's training data.

This charm runs the upstream server as a centrally managed, stateless HTTP endpoint
instead of a per-user local process. It integrates with DataHub over a relation that
provisions a dedicated read-only service account and its access token, with an ingress
provider for a single well-known TLS-terminated URL, and with an identity provider so
that callers authenticate with OAuth 2.1 bearer tokens.

It is intended for data platform teams who want AI-assisted data discovery without
handing long-lived catalog tokens to every workstation. It is a component of the
Canonical Data Mesh solution.