Renta MCP Server

The Renta MCP Server is a Model Context Protocol (MCP) server that connects your AI assistants directly to your Renta workspaces.

It allows you to manage tracking plans, events, properties, ETL sources, and pipelines directly from your favorite IDE or CLI tools.

Available tools and capabilities

The Renta MCP gives an assistant access to your tracking schema, customer data, and ETL pipelines through these tools:

Tool name	Description
renta_assistant	Get the JSON Schema for a specific operation. Returns a detailed schema with required and optional parameters, as well as usage examples. Use before creating or updating objects.
manage_tracking_plans	Manage tracking plans: list, get, create, update, and delete. Each plan is linked to one or more data sources.
manage_events	Manage events within a tracking plan: list, get, create, update, delete, and search by name. Supports track, page, screen, identify, and group types.
bulk_manage_events	Bulk operations on events: create, update, or delete multiple events in a single call. Accepts a JSON array of event objects.
manage_properties	Manage properties (event and user) within a tracking plan: CRUD and search. Supports advanced validation: strings, numbers, enum, regex, arrays, boolean constants, nullable fields.
bulk_manage_properties	Bulk operations on properties: create, update, or delete multiple properties in a single call. Accepts a JSON array of property objects.
tracking_plan_data_quality	Data quality metrics for a tracking plan: events with invalid hits, the properties that failed on an event, and the details of each failed property, for a date range. Drill down step by step: events first, then a property, then its errors.
customer_360	Customer 360 view over Renta CDP: list the trait keys of a tracking plan, find a customer by email, user id, anonymous id, or a set of conditions, and read the customer's activity for a period. Anonymous and identified activity is stitched into one profile.
etl_connectors	List the ETL connectors available through the API, or describe one: its objects, capabilities, and lookups. Call it first, because every other ETL tool takes a source_type from this list.
manage_etl_sources	Manage the accounts ETL pipelines read from: list, get, create, update, patch, and delete. Accounts with a provider sign-in are connected by their owner through an invite link. Deleting a source detaches its pipelines, so it requires confirmation.
manage_etl_invite_links	Invite links that let an account owner connect their account as an ETL source: create, list, get, and revoke. The new source appears in manage_etl_sources once the link is used.
manage_etl_pipelines	Manage ETL pipelines: list, get, create, update, patch, delete, pause, resume, and run. Patch changes only the fields you pass, update replaces the whole configuration.
manage_etl_runs	Work with pipeline runs: list, get, read logs, and cancel. To explain a failed run, find it in the list and read its logs.
etl_describe_fields	List the fields of a source connection or a pipeline body for a connector, with a ready-to-fill JSON template. Fields that take values from a live lookup are marked, so you know when to call etl_field_values.
etl_field_values	Look up the values an ETL field accepts: report types, ad accounts, report fields, destinations. Fixed sets come from the API contract, the rest are read live from the connector. Search narrows large catalogues.
context_layer	Query a Google BigQuery data model through the semantic layer: list models, describe one, sample a field, and run an aggregated query. See the Context Layer page for the full reference.

The context_layer tool is documented separately, because it has its own field syntax, limits, and error codes.

Context Layer

Query your BigQuery data models from an AI assistant.

Integration guides

Here are the step-by-step instructions to add the Renta MCP server to popular AI assistants.

Cursor

Cursor is an AI code editor that natively supports MCP servers.

Open settings. Go to Cursor > Settings… > Cursor Settings.
Navigate to MCP settings. Go to Tools & Integrations > New MCP Server.

Add configuration. Add this configuration to your mcp.json:

mcp.json

{
  "mcpServers": {
    "Renta": {
      "url": "https://mcp.eu.renta.im/mcp"
    }
  }
}

Authenticate. Return to the Tools & Integrations tab and authenticate with Renta.

Claude Desktop

Claude Desktop requires updating its configuration or adding a custom connector.

Navigate to settings. Navigate to claude.ai or open the Claude desktop app.
Add custom connector. Go to Settings > Connectors > Add custom connector.
Configure the integration. Set the name to Renta and URL to https://mcp.eu.renta.im/mcp.
Complete authorization. Complete the OAuth authorization when prompted.

Claude Code

Claude Code is a CLI tool that connects to MCP servers for enhanced context.

Add the MCP server globally. Use the following command to add Renta:
Terminal
```
claude mcp add -t http -s user Renta "https://mcp.eu.renta.im/mcp"
```
Start Claude Code. Run Claude Code in your terminal:
Terminal
```
claude
```
Authenticate with Renta. Initiate the authentication process:
Terminal
```
/mcp
```
Complete authentication. Follow the authentication flow to connect your account.

Gemini CLI

Gemini CLI allows connecting to external MCP servers to assist in your workflow.

Open the configuration settings. Add this to your ~/.gemini/settings.json:

~/.gemini/settings.json

{
  "selectedAuthType": "oauth-personal",
  "mcpServers": {
    "renta": {
      "httpUrl": "https://mcp.eu.renta.im/mcp"
    }
  }
}

Restart the MCP server and authenticate. Run the following command to authenticate:
Terminal
```
gemini/mcp auth renta
```

Codex CLI

Codex CLI supports MCP integration through standard configuration files.

Add the Renta MCP server. Use the following command to add the server:
Terminal
```
codex mcp add renta --url https://mcp.eu.renta.im/mcp
```
Authenticate. Follow the authentication flow when prompted.

Ready to get started?

Build your data pipeline today or get a personalized demo. Start free!

Need help?

Get expert support to ensure your project succeeds. We're here to help!

Feature requests?

Help shape our product! Share your ideas for new features and integrations.