Marmot Data Catalog
- 600 repo stars
- Updated July 23, 2026
- marmotdata/marmot
Marmot Data Catalog is an MCP server that lets agents search your Marmot instance, explore asset lineage, and surface ownership via authenticated remote HTTP tools.
About
Marmot Data Catalog MCP connects AI coding agents to an open-source data catalog you run yourself. developers and small teams shipping data-heavy products use it when agents need grounded answers about datasets, dashboards, jobs, and dependencies instead of guessing from stale READMEs. Registration points the agent at https://{marmot_host}/api/v1/mcp with an X-API-Key header, so every lookup hits your live catalog state. That matters during operate and iterate work: tracing lineage before a migration, finding the owner before a breaking schema change, or scoping impact when you add a new ETL step. It is a protocol bridge to Marmot, not a replacement for dbt, Spark, or your warehouse console—ideal when you already adopted Marmot and want discovery in the same workflow as code changes.
- Remote streamable-http MCP endpoint against your self-hosted Marmot hostname
- Search catalog assets and explore lineage from the agent thread
- Resolve ownership and metadata without leaving Claude Code or Cursor
- Requires Marmot API key from Settings and your instance host URL
- Open-source Marmot repo on GitHub (server schema version 1.0.0)
Marmot Data Catalog by the numbers
- Data as of Jul 28, 2026 (Skillselion catalog sync)
claude mcp add --transport http marmot https://{marmot_host}/api/v1/mcp --header "X-API-Key: YOUR_X_API_KEY"Add your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| repo stars | ★ 600 |
|---|---|
| Transport | HTTP |
| Auth | Required |
| Last updated | July 23, 2026 |
| Repository | marmotdata/marmot ↗ |
What it does
Give your coding agent searchable access to tables, pipelines, lineage, and owners in your Marmot data catalog instance.
Who is it for?
Best when you're already running Marmot and want catalog search and lineage inside Claude Code, Cursor, or Codex.
Skip if: Greenfield projects with no catalog yet, or teams that only need local file search without a Marmot deployment.
What you get
After you register Marmot MCP with host and API key, the agent can query live catalog metadata while you edit pipelines or debug production data issues.
- Agent-callable catalog search and lineage queries against your Marmot host
- Ownership and asset metadata surfaced in the coding session
By the numbers
- MCP server schema version 1.0.0
- Single remote endpoint pattern: /api/v1/mcp on your Marmot host
- Required secret header: X-API-Key
README.md
Marmot
Discover any data asset in seconds. Then let your AI do the same.
The open-source context layer for your AI. Catalog your tables, topics, queues and APIs then expose real metadata to your AI agents.
What is Marmot?
Marmot is an open-source data catalog 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.
Unlike traditional catalogs that require extensive infrastructure and configuration, Marmot ships as a single binary with an intuitive UI, making it easy to deploy and start cataloging in minutes.
Features
- Search everything: Find any data asset in seconds with full-text search plus structured queries, boolean logic and metadata filters.
- Interactive lineage: Trace data flows from source to destination and analyse impact before making changes.
- Metadata-first: Store rich metadata for any asset type, from tables and topics to APIs and dashboards.
- Team collaboration: Assign ownership, document business context and maintain shared glossaries.
- AI-ready: Expose certified context through MCP, the API and the UI.
Deploy
New to Marmot? Follow the Deploy documentation for a guided setup, or try the live demo.
Development
See Local Development for how to get started developing locally.
Community
Join our Discord community for help, feedback and updates on new features.
Contributing
All types of contributions are encouraged and valued!
- Report bugs or suggest features via GitHub Issues
- Improve documentation
- Build new plugins for data sources
Before contributing, please check out the Contributing Guide.
FAQ
What is Marmot?
Marmot is an open-source data catalog for teams who want powerful data discovery without enterprise complexity. It catalogs every data asset (tables, topics, queues, APIs), enriches it with context, and makes it accessible to both your team and AI tools. Unlike traditional catalogs requiring extensive infrastructure, Marmot ships as a single binary with an intuitive UI.
Why Use Marmot?
| Feature | Benefit |
|---|---|
| Search Everything | Find any data asset in seconds with full-text search, structured queries, boolean logic, and metadata filters |
| Interactive Lineage | Trace data flows from source to destination, analyze impact before making changes |
| Metadata-First | Store rich metadata for any asset type (tables, topics, APIs, dashboards) |
| Team Collaboration | Assign ownership, document business context, maintain shared glossaries |
| AI-Ready | Expose certified context through MCP, API, and UI |
| Single Binary | No complex infrastructure required, deploy in minutes |
How Does Marmot Integrate with AI Agents?
Marmot exposes certified context through MCP (Model Context Protocol), enabling AI agents to query real metadata about your data assets. This allows AI tools to understand your data landscape, trace lineage, and make informed decisions based on actual metadata rather than guessing.
What Data Sources Can I Catalog?
Marmot supports cataloging various data asset types through its plugin system:
- Tables (databases, data warehouses)
- Topics (message queues, event streams)
- Queues (job queues, messaging systems)
- APIs (REST, GraphQL, internal services)
- Dashboards (visualization tools, BI platforms)
How Do I Deploy Marmot?
Quick Start Options:
| Method | Description |
|---|---|
| Documentation Guide | Follow the Deploy guide for step-by-step setup |
| Live Demo | Try the live demo before deploying |
| Single Binary | Download and run the single binary for your platform |
Is Marmot Free?
Yes! Marmot is open-source software licensed under the MIT License. You can use, modify, and distribute it freely. Self-hosting is completely free with no licensing costs.
How Can I Contribute?
All contributions are welcome:
- Report bugs or suggest features via GitHub Issues
- Improve documentation (README, guides, API docs)
- Build new plugins for additional data sources
- Check the Contributing Guide before contributing
Where Can I Get Help?
| Resource | Link |
|---|---|
| Documentation | marmotdata.io/docs |
| Discord Community | Join Discord |
| GitHub Issues | Report issues |
| Live Demo | Try demo |
License
Marmot is open-source software licensed under the MIT License.
Recommended MCP Servers
1ClickReport40 tools for marketing analytics, campaign management, and AI monitoring across 6 platforms.
222wcnm BilistalkermcpTrack Bilibili creators and get the latest updates on videos, dynamics, and articles.7
ABMeterFeature flagging and A/B testing platform with AI-first experimentation workflows.
Abs McpQuery the ABS in plain English — CPI, labour force, building approvals, migration, GDP.
AdAdvisor MCP ServerQuery Meta Ads performance data — accounts, campaigns, ad sets, ads, metrics & settings.
ADCIRC Model Setup MCP ServerADCIRC model configuration debugging, parameter lookup, and validation5How it compares
MCP integration to a data catalog, not an in-repo codebase intelligence skill like Sourcebook.
FAQ
Who is Marmot Data Catalog MCP for?
Developers and small teams who self-host Marmot and want agents to discover datasets, lineage, and owners without switching to the web UI.
When should I use Marmot Data Catalog MCP?
Use it during operate and iterate work when you need accurate asset search, downstream lineage, or owner lookup before changing pipelines or schemas.
How do I add Marmot Data Catalog MCP to my agent?
Add the remote MCP server URL https://{marmot_host}/api/v1/mcp, set the marmot_host variable to your instance, and supply X-API-Key from Marmot Settings > API Keys.