MCP Data ServerThe bridge between AI agents and cloud-native data

An open Model Context Protocol server that grounds coding agents in STAC metadata and confines them to validated cloud-native engines — so any LLM client can query terabyte-scale data over S3 without downloading it, misreading it, or silently failing at scale.

Quick Start

The bigger picture

Browse Datasets

Grounded in STAC metadata

The agent browses a standard STAC catalog to find the right dataset and read its column schema before writing a query — so it interprets the data correctly instead of guessing.

Validated cloud-native engines

Queries run against Parquet on S3 with DuckDB — fast, out-of-core, columnar. The agent reaches for streaming engines instead of in-memory libraries that silently break at scale.

Runs local or at scale

Run it on your own hardware for sensitive data, or on autoscaling Kubernetes for terabyte workloads. The same server, the same tools, either way.

Drivable by small open models

The query guidance is injected at call time, so even compact, locally-run open models can drive the workflow — reducing dependence on closed models and keeping data on your hardware.

Private data ready

Pass S3 credentials per call to query private buckets alongside public data. Credentials are request-scoped, never logged, and never shared between clients.

Part of a larger effort

One of three open-source components — with data-workflows and jupyter-geoagent — that together make the cloud-native stack reachable by the AI tools researchers already use.