# Knowledge Catalog MCP > Google Cloud managed MCP server for Knowledge Catalog. Exact tool names and accessible operations depend on the published server reference and configured Google Cloud permissions. Canonical: https://www.mcpfolio.com/servers/google-cloud-knowledge-catalog ## Catalog record - Publisher: Google Cloud - Categories: search, cloud-service ## Website https://docs.cloud.google.com/dataplex/docs/reference/mcp ## Transports Streamable HTTP ## Authentication Google Cloud authentication: OAuth 2.0 using Application Default Credentials (ADC) in supported Google clients, or a client OAuth ID/secret or bearer token in compatible third-party clients. The principal needs mcp.tools.call plus the underlying product permissions; API key authentication is only available for services that explicitly support it. ## Tools ### search_entries Input schema is not included in the available catalog record. ### get_data_product Input schema is not included in the available catalog record. ### get_data_asset Input schema is not included in the available catalog record. ### list_data_products Input schema is not included in the available catalog record. ### list_data_assets Input schema is not included in the available catalog record. ### create_data_product Input schema is not included in the available catalog record. ### update_data_product Input schema is not included in the available catalog record. ### create_data_asset Input schema is not included in the available catalog record. ### update_data_asset Input schema is not included in the available catalog record. ### get_operation Input schema is not included in the available catalog record. ### lookup_context Input schema is not included in the available catalog record. ### lookup_entry Input schema is not included in the available catalog record. ### list_data_product_aspect_types Input schema is not included in the available catalog record. ### update_data_product_aspects Input schema is not included in the available catalog record. ### search_lineage Input schema is not included in the available catalog record. ## Sources - [official_tool_reference](https://docs.cloud.google.com/dataplex/docs/reference/mcp) - [official_server_catalog](https://docs.cloud.google.com/mcp/supported-products) - [official_setup_guide](https://docs.cloud.google.com/mcp/configure-mcp-ai-application) - [official_authentication_guide](https://docs.cloud.google.com/mcp/authenticate-mcp) ## Client setup instructions ### Cursor Status: ready to configure. Configuration: ```json { "mcpServers": { "google-cloud-knowledge-catalog": { "url": "https://dataplex.googleapis.com/mcp" } } } ``` Prerequisites: - A Google Cloud project and access to the target product/resource. - Enable this MCP service and its underlying Google Cloud API as required by the product guide. - An MCP client that supports Streamable HTTP. - An identity with the MCP Tool User role (roles/mcp.toolUser) or equivalent mcp.tools.call permission, plus permissions for the underlying product operations. Authentication: - If OAuth is required, connect through Cursor's MCP tools settings and finish the browser sign-in yourself. For credential-based servers, store credentials in a local environment variable or the client's supported secret facility; do not paste secrets into chat. Setup steps: - Choose the correct global or regional endpoint for your data location; this catalog lists the documented endpoint. - In the Google Cloud project, enable the MCP service and the underlying product API if the product guide requires it. - Authenticate the client with Google Cloud ADC (simplest with Antigravity/Gemini CLI) or configure OAuth 2.0 credentials in a compatible client. - Grant the connecting identity roles/mcp.toolUser and only the product IAM roles needed for the intended operations. - Add dataplex.googleapis.com to the client as a Streamable HTTP MCP server and complete sign-in. - Refresh the client tool list and check that the tools match the product reference; availability may depend on permissions and selected toolset. Reload: - Open Cursor Settings → Tools & MCP; restart or reconnect the server if Cursor does not pick up the saved project configuration. Verify the connection: - Open Cursor Settings → Tools & MCP and confirm the server is enabled with its tools listed. Ask for a harmless read-only operation only after authentication. Official client guide: [Cursor](https://docs.cloud.google.com/dataplex/docs/reference/mcp) ### VS Code Status: ready to configure. Configuration: ```json { "mcpServers": { "google-cloud-knowledge-catalog": { "type": "http", "url": "https://dataplex.googleapis.com/mcp" } } } ``` Prerequisites: - A Google Cloud project and access to the target product/resource. - Enable this MCP service and its underlying Google Cloud API as required by the product guide. - An MCP client that supports Streamable HTTP. - An identity with the MCP Tool User role (roles/mcp.toolUser) or equivalent mcp.tools.call permission, plus permissions for the underlying product operations. Authentication: - For OAuth, use VS Code's MCP server controls and complete the browser sign-in yourself. For credential-based servers, use a supported local environment variable/input mechanism; never write a real secret into a tracked workspace file. Setup steps: - Choose the correct global or regional endpoint for your data location; this catalog lists the documented endpoint. - In the Google Cloud project, enable the MCP service and the underlying product API if the product guide requires it. - Authenticate the client with Google Cloud ADC (simplest with Antigravity/Gemini CLI) or configure OAuth 2.0 credentials in a compatible client. - Grant the connecting identity roles/mcp.toolUser and only the product IAM roles needed for the intended operations. - Add dataplex.googleapis.com to the client as a Streamable HTTP MCP server and complete sign-in. - Refresh the client tool list and check that the tools match the product reference; availability may depend on permissions and selected toolset. Reload: - Run `MCP: List Servers` from the Command Palette and use its Start/Restart action, or reload the VS Code window if needed. Verify the connection: - Run `MCP: List Servers`; confirm the server is running and its tools appear in Copilot Chat's Agent mode. Official client guide: [VS Code](https://docs.cloud.google.com/dataplex/docs/reference/mcp) ### Claude Desktop Status: ready to configure. Configuration: ```json Remote connector URL: https://dataplex.googleapis.com/mcp Add it in Claude Desktop → Settings → Connectors → Add custom connector. ``` Prerequisites: - A Google Cloud project and access to the target product/resource. - Enable this MCP service and its underlying Google Cloud API as required by the product guide. - An MCP client that supports Streamable HTTP. - An identity with the MCP Tool User role (roles/mcp.toolUser) or equivalent mcp.tools.call permission, plus permissions for the underlying product operations. Authentication: - Remote connector: review Claude's detected authentication and complete OAuth in its browser flow, or enter a required API credential in the connector's request-header UI. Local extensions may request settings in Claude Desktop; enter secrets only into its secure settings UI. Setup steps: - Choose the correct global or regional endpoint for your data location; this catalog lists the documented endpoint. - In the Google Cloud project, enable the MCP service and the underlying product API if the product guide requires it. - Authenticate the client with Google Cloud ADC (simplest with Antigravity/Gemini CLI) or configure OAuth 2.0 credentials in a compatible client. - Grant the connecting identity roles/mcp.toolUser and only the product IAM roles needed for the intended operations. - Add dataplex.googleapis.com to the client as a Streamable HTTP MCP server and complete sign-in. - Refresh the client tool list and check that the tools match the product reference; availability may depend on permissions and selected toolset. Reload: - For a remote connector, enable it from Claude's Connectors UI. For a newly installed local desktop extension, restart Claude Desktop if its tools do not appear. Verify the connection: - Open the Connectors/tools picker and confirm the server and its tools are available. Check Developer settings and logs for local extension failures. Official client guide: [Claude Desktop](https://docs.cloud.google.com/dataplex/docs/reference/mcp) ### Claude Code Status: ready to configure. Configuration: ```text claude mcp add --transport http google-cloud-knowledge-catalog --scope project "https://dataplex.googleapis.com/mcp" ``` Prerequisites: - A Google Cloud project and access to the target product/resource. - Enable this MCP service and its underlying Google Cloud API as required by the product guide. - An MCP client that supports Streamable HTTP. - An identity with the MCP Tool User role (roles/mcp.toolUser) or equivalent mcp.tools.call permission, plus permissions for the underlying product operations. Authentication: - For OAuth, run 'claude /mcp' in a regular terminal and authenticate google-cloud-knowledge-catalog if prompted. Complete the browser sign-in yourself; do not paste credentials into chat. Setup steps: - Choose the correct global or regional endpoint for your data location; this catalog lists the documented endpoint. - In the Google Cloud project, enable the MCP service and the underlying product API if the product guide requires it. - Authenticate the client with Google Cloud ADC (simplest with Antigravity/Gemini CLI) or configure OAuth 2.0 credentials in a compatible client. - Grant the connecting identity roles/mcp.toolUser and only the product IAM roles needed for the intended operations. - Add dataplex.googleapis.com to the client as a Streamable HTTP MCP server and complete sign-in. - Refresh the client tool list and check that the tools match the product reference; availability may depend on permissions and selected toolset. Reload: - Start or restart the Claude Code session in the project if it was already open. Review and approve a new project-scoped server when Claude Code asks. Verify the connection: - Run 'claude mcp get google-cloud-knowledge-catalog' or 'claude mcp list'; in an interactive session, use '/mcp'. Confirm the status is connected and tools are listed. A configured entry alone is not proof of connection. Official client guide: [Claude Code](https://docs.cloud.google.com/dataplex/docs/reference/mcp) ### Gemini CLI Status: ready to configure. Configuration: ```text gemini mcp add --transport http google-cloud-knowledge-catalog "https://dataplex.googleapis.com/mcp" ``` Prerequisites: - A Google Cloud project and access to the target product/resource. - Enable this MCP service and its underlying Google Cloud API as required by the product guide. - An MCP client that supports Streamable HTTP. - An identity with the MCP Tool User role (roles/mcp.toolUser) or equivalent mcp.tools.call permission, plus permissions for the underlying product operations. Authentication: - For OAuth, run '/mcp auth google-cloud-knowledge-catalog' inside Gemini CLI and complete the browser flow yourself. Setup steps: - Choose the correct global or regional endpoint for your data location; this catalog lists the documented endpoint. - In the Google Cloud project, enable the MCP service and the underlying product API if the product guide requires it. - Authenticate the client with Google Cloud ADC (simplest with Antigravity/Gemini CLI) or configure OAuth 2.0 credentials in a compatible client. - Grant the connecting identity roles/mcp.toolUser and only the product IAM roles needed for the intended operations. - Add dataplex.googleapis.com to the client as a Streamable HTTP MCP server and complete sign-in. - Refresh the client tool list and check that the tools match the product reference; availability may depend on permissions and selected toolset. Reload: - Gemini CLI reads its settings on startup. Restart the CLI after direct file edits; its `mcp add` command writes the configuration. Verify the connection: - Run 'gemini mcp list' and confirm google-cloud-knowledge-catalog reports Connected. For local stdio servers, trust the current folder if Gemini CLI requires it. Official client guide: [Gemini CLI](https://docs.cloud.google.com/dataplex/docs/reference/mcp) ### Codex CLI Status: ready to configure. Configuration: ```text codex mcp add google-cloud-knowledge-catalog --url "https://dataplex.googleapis.com/mcp" ``` Prerequisites: - A Google Cloud project and access to the target product/resource. - Enable this MCP service and its underlying Google Cloud API as required by the product guide. - An MCP client that supports Streamable HTTP. - An identity with the MCP Tool User role (roles/mcp.toolUser) or equivalent mcp.tools.call permission, plus permissions for the underlying product operations. Authentication: - For OAuth, run 'codex mcp login google-cloud-knowledge-catalog' and complete the browser flow yourself. Do not paste credentials into chat. Setup steps: - Choose the correct global or regional endpoint for your data location; this catalog lists the documented endpoint. - In the Google Cloud project, enable the MCP service and the underlying product API if the product guide requires it. - Authenticate the client with Google Cloud ADC (simplest with Antigravity/Gemini CLI) or configure OAuth 2.0 credentials in a compatible client. - Grant the connecting identity roles/mcp.toolUser and only the product IAM roles needed for the intended operations. - Add dataplex.googleapis.com to the client as a Streamable HTTP MCP server and complete sign-in. - Refresh the client tool list and check that the tools match the product reference; availability may depend on permissions and selected toolset. Reload: - Restart an already-open Codex session or IDE extension if the server does not appear; the IDE extension offers a Restart action. Verify the connection: - Run 'codex mcp list' or '/mcp' in the Codex TUI. Confirm the server is enabled/connected and its tools are available. Official client guide: [Codex CLI](https://docs.cloud.google.com/dataplex/docs/reference/mcp) ### Windsurf Status: ready to configure. Configuration: ```json { "mcpServers": { "google-cloud-knowledge-catalog": { "serverUrl": "https://dataplex.googleapis.com/mcp" } } } ``` Prerequisites: - A Google Cloud project and access to the target product/resource. - Enable this MCP service and its underlying Google Cloud API as required by the product guide. - An MCP client that supports Streamable HTTP. - An identity with the MCP Tool User role (roles/mcp.toolUser) or equivalent mcp.tools.call permission, plus permissions for the underlying product operations. Authentication: - OAuth is supported for remote and stdio server transports; complete any browser sign-in from Cascade. For a fixed credential, store it using documented environment interpolation rather than a literal secret in JSON. Setup steps: - Choose the correct global or regional endpoint for your data location; this catalog lists the documented endpoint. - In the Google Cloud project, enable the MCP service and the underlying product API if the product guide requires it. - Authenticate the client with Google Cloud ADC (simplest with Antigravity/Gemini CLI) or configure OAuth 2.0 credentials in a compatible client. - Grant the connecting identity roles/mcp.toolUser and only the product IAM roles needed for the intended operations. - Add dataplex.googleapis.com to the client as a Streamable HTTP MCP server and complete sign-in. - Refresh the client tool list and check that the tools match the product reference; availability may depend on permissions and selected toolset. Reload: - Save `mcp_config.json`, then restart or reconnect the server from Cascade's MCP controls. Verify the connection: - Open Cascade's MCP panel and confirm the server is enabled and lists tools. Note: this documentation applies to legacy Cascade; newer Devin Local setup may differ. Official client guide: [Windsurf](https://docs.cloud.google.com/dataplex/docs/reference/mcp) ### ChatGPT Status: ready to configure. Configuration: ```json Remote MCP endpoint: https://dataplex.googleapis.com/mcp Add it in ChatGPT web → Settings/Workspace Settings → Apps → Create. ``` Prerequisites: - A Google Cloud project and access to the target product/resource. - Enable this MCP service and its underlying Google Cloud API as required by the product guide. - An MCP client that supports Streamable HTTP. - An identity with the MCP Tool User role (roles/mcp.toolUser) or equivalent mcp.tools.call permission, plus permissions for the underlying product operations. Authentication: - If OAuth is configured, complete the authorization prompt while ChatGPT scans tools. Workspace admin/developer permissions may be required. Do not paste credentials into chat. Setup steps: - Choose the correct global or regional endpoint for your data location; this catalog lists the documented endpoint. - In the Google Cloud project, enable the MCP service and the underlying product API if the product guide requires it. - Authenticate the client with Google Cloud ADC (simplest with Antigravity/Gemini CLI) or configure OAuth 2.0 credentials in a compatible client. - Grant the connecting identity roles/mcp.toolUser and only the product IAM roles needed for the intended operations. - Add dataplex.googleapis.com to the client as a Streamable HTTP MCP server and complete sign-in. - Refresh the client tool list and check that the tools match the product reference; availability may depend on permissions and selected toolset. Reload: - Save/Create the app draft, then open a new chat and select the app from the tools menu. Admin approval or publishing may be required for workspace use. Verify the connection: - Confirm the app appears as a draft or enabled app, its tool scan completes, and the tools are available in a chat. ChatGPT web cannot connect directly to local stdio servers. Official client guide: [ChatGPT](https://docs.cloud.google.com/dataplex/docs/reference/mcp) ### Manual Status: ready to configure. Configuration: ```json { "mcpServers": { "google-cloud-knowledge-catalog": { "url": "https://dataplex.googleapis.com/mcp" } } } ``` Prerequisites: - A Google Cloud project and access to the target product/resource. - Enable this MCP service and its underlying Google Cloud API as required by the product guide. - An MCP client that supports Streamable HTTP. - An identity with the MCP Tool User role (roles/mcp.toolUser) or equivalent mcp.tools.call permission, plus permissions for the underlying product operations. Authentication: - Follow the selected client's official OAuth flow or store a required credential in its local secret/environment mechanism. Never paste secrets into chat or commit them. Setup steps: - Choose the correct global or regional endpoint for your data location; this catalog lists the documented endpoint. - In the Google Cloud project, enable the MCP service and the underlying product API if the product guide requires it. - Authenticate the client with Google Cloud ADC (simplest with Antigravity/Gemini CLI) or configure OAuth 2.0 credentials in a compatible client. - Grant the connecting identity roles/mcp.toolUser and only the product IAM roles needed for the intended operations. - Add dataplex.googleapis.com to the client as a Streamable HTTP MCP server and complete sign-in. - Refresh the client tool list and check that the tools match the product reference; availability may depend on permissions and selected toolset. Reload: - Restart/reload the selected client only if its documentation requires it. Verify the connection: - Use the selected client's MCP status view and verify the server exposes tools; do not report connected based only on a saved config. Official client guide: [Manual](https://docs.cloud.google.com/dataplex/docs/reference/mcp)