Launching today

Marmot
AI-native data catalog with search, lineage and MCP
40 followers
AI-native data catalog with search, lineage and MCP
40 followers
Marmot is an open-source data catalog designed for teams who want powerful data discovery without enterprise complexity. Catalog every data asset, enrich it with the context that matters and make it accessible to your team and your AI tools.







Marmot
Hey everyone, I'm Charlie - I built Marmot because data catalogs solve a simple problem: what data do we have and where does it live? But most catalogs will then ask you to run Elasticsearch, Kafka and a graph database just to find out. Data catalogs shouldn't need an entire platform team to run them.
Now your AI agents need to answer these same questions - without valid context, they hallucinate. Marmot gives them a single source of truth to query.
Marmot is a single binary and a Postgres database, deployable in minutes. It currently has:
25+ plugins (and growing) including - dbt, Kafka, S3, Trino, Iceberg, PostgreSQL.
Full data lineage across your business
Built-in MCP server to give LLMs context around your data
100% free and open-source.
Try it: demo.marmotdata.io
Star us: github.com/marmotdata/marmot
Try it out and let me know what you think!
I’m curious how easy it is for me to deploy and start using quickly without needing a dedicated data governance team. A quick-start guide or demo would really help me evaluate this faster.
Marmot
Hey @morgan_nabors, there's quick start guide to quickly deploy with Docker Compose. There's also a live Demo Site to evaluate without needing to deploy anything at all! https://demo.marmotdata.io/
I'm planning to build a free hosted tier with some usage limits in the coming months to help teams evaluate whether Marmot works for them without committing. Let me know how you get on! Happy to help out where I can
I’ve been thinking a lot about how teams can bridge the gap between data and AI and this feels like a step in that direction. For me, the biggest challenge is not just finding data but trusting it.
Marmot
@martha_s_bako It's definitely becoming a more prevalent problem, especially as engineers start relying on AI tools in their daily workflows. Having access to valid, trustworthy data instantly speeds up cycle time for engineers and prevents AI from making confident-sounding bad decisions.