# Marmot Data Catalog [Health: Active]

**Category:** 📊 Data Platforms  
**Repository:** https://github.com/marmotdata/marmot  
**GitHub Stars:** 623  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/marmot-data-catalog

## Description
Open-source data catalog. Search assets, explore lineage, and find ownership.

## Claude Desktop Quick Installation
No confirmed install details yet, so no config is published. Follow the setup steps in the repository above.

## Documentation

## What Marmot Data Catalog MCP server does

Marmot Data Catalog MCP server gives AI agents access to metadata maintained in the Marmot open-source data catalog. The catalog is designed to describe a broad set of data assets, including tables, message topics, queues, APIs, and dashboards.

Through MCP, agents can use catalog context to find relevant assets, understand how data moves between sources and destinations, and identify the ownership or business information recorded by a team. Marmot also provides a user interface and API, but this listing focuses on its MCP exposure for agent access.

The catalog supports full-text search, structured queries, boolean logic, and metadata filters. It can also hold ownership assignments, business descriptions, and shared glossary information. These capabilities make it suitable for questions such as locating a dataset, checking related assets, or examining the possible impact of a change.

## How it works

Marmot acts as a metadata layer between data assets and the AI tools that need to understand them. Plugins collect or represent information about supported asset types, while the catalog stores the resulting metadata and relationships. AI agents access certified catalog context through MCP rather than querying each underlying system directly.

Lineage information lets an agent follow flows from an origin to a destination and review connected assets. Search and metadata filters narrow the catalog to relevant results. The quality of agent responses depends on the assets, relationships, ownership details, and other context recorded in the Marmot instance.

## Setup and configuration

Marmot is distributed as a single binary and can be deployed using the project’s deployment documentation.

After deployment, teams can catalog data assets through Marmot’s plugin system and expose the catalog through MCP. The project also provides documentation for local development, a deployment guide, and community support through Discord and GitHub Issues.

## Tools and capabilities

The Marmot Data Catalog MCP server supports agent workflows involving:

- Searching cataloged assets with text search, structured queries, boolean logic, and metadata filters.
- Reviewing metadata for tables, topics, queues, APIs, dashboards, and other cataloged assets.
- Exploring lineage between source and destination assets.
- Investigating ownership and documented business context.
- Using shared glossary information maintained by the catalog.

## Limitations and notes

It also does not list every available catalog plugin or provide details about how individual data sources are connected.

Marmot is open-source software under the MIT License. Teams should consult the deployment and product documentation before configuring a production instance, especially when they need source-specific ingestion details or a precise MCP setup procedure.

_Full upstream README: https://allmcps.com/mcp/marmot-data-catalog/readme_