CLIO Ndp
Search and discover National Data Platform datasets from your agent while scoping a data-heavy product or research feature.
Overview
CLIO Ndp is an Idea-phase MCP server that connects agents to National Data Platform dataset search and discovery across multiple sources.
What is this MCP server?
- National Data Platform (NDP) search and dataset discovery via MCP
- Multi-catalog dataset exploration suited to early product research
- CLIO Ndp in clio-kit 2.2.3 with stdio MCP transport
- Part of IOWarp CLIO toolkit alongside scientific infrastructure MCP servers
- Keeps provenance-oriented discovery inside the agent instead of tab hopping
- CLIO Ndp version 2.2.3
- stdio transport
- clio-kit PyPI package
Community signal: 25 GitHub stars.
What problem does it solve?
Jumping between NDP portals and copy-pasting dataset IDs breaks flow when an agent is helping you evaluate whether an idea has viable public data.
Who is it for?
Indie builders exploring data products, research tooling, or ML demos that depend on discovering NDP-hosted or federated datasets early.
Skip if: Teams with no data-discovery need, or products that use only private first-party data with no public catalog research.
What do I get? / Deliverables
After registration, your agent can query NDP-oriented discovery tools so research summaries and scope docs reference concrete datasets faster.
- Agent-callable NDP search and discovery tools
- Faster dataset shortlists for idea-phase specs and prototypes
- Stdio MCP integration via CLIO Ndp
Recommended MCP Servers
Journey fit
Dataset discovery belongs in the idea phase when you are still choosing what signals, training data, or public records your build should rely on. Research is the right shelf because NDP MCP is optimized for finding and comparing datasets across platforms before you commit to a schema or pipeline.
How it compares
Public dataset discovery MCP, not a warehouse ETL skill or proprietary analytics connector.
Common Questions / FAQ
Who is io.github.iowarp/ndp-mcp for?
It is for builders and researchers who use the National Data Platform and want Claude Code, Cursor, or similar agents to help search and discover datasets via MCP.
When should I use io.github.iowarp/ndp-mcp?
Use it in the idea and validate research window when you are comparing datasets, framing a problem statement, or drafting a prototype data plan.
How do I add io.github.iowarp/ndp-mcp to my agent?
Install clio-kit from PyPI, configure the Ndp MCP server (stdio) in your agent’s MCP file, and ensure any NDP access or network requirements your deployment needs are satisfied.