Repolode
Submit
marmotdata

marmotdata/marmot

Open-source data catalog that indexes tables, topics, queues and APIs and exposes their metadata to AI agents via MCP.

Open-source alternative to:AlationCollibra

View on GitHubWebsite
Stars
634
Forks
33
Open issues
43
Last commit
2 days ago

What is it?

Marmot is an open-source data catalog for teams that want data discovery without the infrastructure overhead of traditional enterprise catalogs. It catalogs tables, topics, queues, APIs and dashboards, lets teams attach ownership, business context and shared glossaries, and provides full-text and structured search with metadata filters.

It includes interactive lineage for tracing data flows from source to destination and assessing the impact of a change. The project ships as a single binary with a web UI, and its plugin system is the route for adding new data source types.

Marmot also exposes its catalog through MCP (Model Context Protocol), the API and the UI, so AI agents can query real metadata about your data assets rather than guessing at your data landscape.

Key features

  • 01Full-text search with structured queries, boolean logic and metadata filters.
  • 02Interactive lineage graph showing data flows and downstream impact.
  • 03Stores rich metadata for tables, topics, queues, APIs and dashboards.
  • 04Ownership assignment, business documentation and shared glossaries.
  • 05Exposes catalog context to AI agents through MCP, the API and the UI.
  • 06Ships as a single binary.

Use cases

  • Helping data teams find the right table or topic across many systems.
  • Tracing where a dataset comes from and what breaks if it changes.
  • Giving AI agents accurate metadata about company data assets through MCP.
  • Documenting data ownership and business definitions in one shared place.

Quick start

  1. 1

    Follow the Deploy guide in the documentation for a guided setup

  2. 2

    Download and run the single binary for your platform

  3. 3

    Follow the local development guide to get started developing locally

Things to know

  • The README gives no concrete install command, so setup requires following the external documentation.
  • Early-stage project with 628 stars; the README does not state a stability or production-readiness level.

In collections

MCP servers

Topics

#data-catalog#data-discovery#data-lineage#metadata#mcp#data-governance#bigdata#data-collaboration#data-exploration#data-observability

FAQ

What is marmot?
Open-source data catalog that indexes tables, topics, queues and APIs and exposes their metadata to AI agents via MCP. Marmot is an open-source data catalog for teams that want data discovery without the infrastructure overhead of traditional enterprise catalogs. It catalogs tables, topics, queues, APIs and dashboards, lets teams attach ownership, business context and shared glossaries, and provides full-text and structured search with metadata filters.
Is marmot free to use? What license does it have?
marmot is open source under the MIT license. Check the license terms before using it in commercial products.
Is marmot still maintained?
Actively maintained. The last commit to marmot was on Nov 5, 2026.
What can I use marmot for?
Helping data teams find the right table or topic across many systems. Tracing where a dataset comes from and what breaks if it changes. Giving AI agents accurate metadata about company data assets through MCP. Documenting data ownership and business definitions in one shared place.
How do I install marmot?
Follow the Deploy guide in the documentation for a guided setup Download and run the single binary for your platform Follow the local development guide to get started developing locally
Where can I get marmot?
The source code and installation instructions are on GitHub: https://github.com/marmotdata/marmot

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects