nasa-power-mcp-server

Overview

NASA POWER historical solar and meteorological observations for renewable energy, sustainable buildings and agroclimatology. Retrieve point time series and climatology, inspect native-grid regions, and compare labeled locations with source units and provenance. No NASA account or API key is required.

Tools

Eight tools are enabled by default; NASA_POWER_DATAFRAME_DROP_ENABLED=true enables the ninth.

Related MCP server: @cyanheads/eia-energy-mcp-server

Capability reference

nasa_power_get_timeseries tool

• latitude / longitude, temporalResolution (hourly, daily, monthly), and an inclusive startDate / endDate or startYear / endYear window. selection chooses a named preset or explicit parameters: at most 15 effective hourly codes or 20 otherwise, including corrections.

• community (RE, SB, AG; default RE), units (metric, imperial) and timeStandard (LST, UTC) retain the returned source units and time labels. Optional siteElevation adds PSC; paired windElevation (10–300 meters) and windSurface (grass, airportice) add WSC.

• Returns request, metadata, parameters, expansions, omittedParameters, summaries, rows, exact rowCount and nullable dataset. Optional canvas_id reuses a workspace when staging is needed.

nasa_power_get_climatology tool

• Point coordinates and the same parameter selection and corrections; optionally supply both startYear and endYear, otherwise use the source product range. No temporalResolution input is needed.

• Complete monthly and annual records remain inline; the climatology alias SI_EF_TILTED_SURFACE preserves returned numeric angles and orientation text through expansions and parameter metadata.

nasa_power_get_region tool

• One parameter, ordered latitudeMin / latitudeMax and longitudeMin / longitudeMax, each spanning 2–10 degrees; split antimeridian regions. Supports daily, monthly and climatology windows, with common community, units and time settings.

• Returns cell-tagged observations, cellCount and per-period cellSummaries. Means weight each returned grid cell equally; point correction fields and hourly regions are unsupported. Optional canvas_id reuses a staging workspace.

nasa_power_compare_locations tool

• 2–10 locations, each with coordinates and a label of 1–120 characters, plus shared selection, corrections and window. Defaults to climatology; supports hourly, daily and monthly comparisons too.

• Location-tagged rows, per-site metadata and summaries, seasonal buckets, partial and failed preserve operational failures alongside successes. possibleSharedGridCell flags nearby points without asserting exact grid identity.

• Optional sort takes a numeric selected parameter and direction (asc, desc); sorting requires known identical units. Optional canvas_id reuses a workspace. All-site failure returns all_locations_failed.

nasa_power_search_parameters tool

• community (default RE), temporalResolution (default daily), optional filter (up to 256 characters), offset and limit 1–100 (default 20).

• Case-insensitive tokens filter the complete dictionary before sorted pagination; blank filter browses it. entries, total, hasMore and truncated describe the filtered page.

nasa_power_list_reference tool

• topic: presets (default), communities, coverage, missing_data, wind_surfaces or dataframes.

• Returns typed entries, guidance, documentationUrls and verifiedOn. Presets are solar_pv, wind_turbine, building_hvac, agroclimatology and water_balance; building_hvac is unavailable hourly.

nasa_power_dataframe_describe tool

• Required canvas_id from a producing response; optional table limits inspection to one table.

• Returns current canvas/table expiry, actual SQL columns with source units and original scientific provenance. Only tables in the supplied canvas are visible.

nasa_power_dataframe_query tool

• Required canvas_id and one read-only SELECT in sql (up to 16,384 characters); rowLimit 1–1,000 (default 100). Joins and aggregates operate on tables shown by describe; files, writes and system catalogs are denied.

• Returns columns, rows, rowCount, shownCount, truncated and countSemantics (exact or materialized_prefix). SQL has a 10-second deadline; the complete structured and rendered outputs each must fit 40,000 characters.

nasa_power_dataframe_drop tool

• Required canvas_id and exact table; enabled only by NASA_POWER_DATAFRAME_DROP_ENABLED=true.

• Returns dropped: false for an absent table in a live canvas. Expired or unknown canvases return canvas_not_found.

Features

Built on @cyanheads/mcp-ts-core: stdio and Streamable HTTP transports, pluggable auth (none / jwt / oauth), swappable storage (in-memory, filesystem, Supabase, Cloudflare KV/R2/D1), structured logging with optional OpenTelemetry tracing.

POWER-specific:

• Historical gridded observations, rather than site measurements or forecasts. Meteorological coverage starts in 1981, solar in 1984 and hourly in 2001; source configuration and publication latency further constrain availability.

• One active upstream request, starts at least two seconds apart, bounded retries and a 55-second overall call deadline. NASA publishes no numeric request allowance. References cache for 24 hours and observations for 10 minutes.

• Source fill values become null; zero remains a measurement. Monthly annual keys and climatology ANN remain distinct from monthly observations. LST labels have no UTC offset.

Agent-friendly output:

• Scientific provenance — source API version, dataset identifiers, endpoint and access time accompany results; preserve returned units and attribute NASA POWER.

• Explicit gaps — omittedParameters distinguishes an absent series from fill values; arithmetic observation means and calendar-season buckets are not climate normals, annual yields or water balances.

• Complete datasets — inline outputs each fit 40,000 characters. Larger datasets stage all rows with a preview of at most 20 rows and a dataset handle; unsupported or oversized requests fail with narrowing guidance instead of returning a capped source prefix.

Staging requires DuckDB and supports at most 250,000 rows, 64 parameter columns and 32 MiB of serialized rows per dataset. Canvases are process-local capabilities: pass the returned token explicitly and route later calls to the same process. Tables expire after 30 minutes, canvases after one idle hour with a six-hour absolute cap; limits are eight tables per canvas, four canvases per tenant and eight per process. Engine memory is 256 MiB per canvas, not a total RSS or temporary-disk quota.

Getting started

Self-Hosted / Local

Add this to your MCP client configuration:

{
  "mcpServers": {
    "nasa-power-mcp-server": {
      "type": "stdio",
      "command": "bunx",
      "args": ["@cyanheads/nasa-power-mcp-server@latest"],
      "env": { "MCP_TRANSPORT_TYPE": "stdio" }
    }
  }
}

Or with npx (no Bun required):

{
  "mcpServers": {
    "nasa-power-mcp-server": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@cyanheads/nasa-power-mcp-server@latest"],
      "env": { "MCP_TRANSPORT_TYPE": "stdio" }
    }
  }
}

Or with Docker:

{
  "mcpServers": {
    "nasa-power-mcp-server": {
      "type": "stdio",
      "command": "docker",
      "args": ["run", "-i", "--rm", "-e", "MCP_TRANSPORT_TYPE=stdio", "ghcr.io/cyanheads/nasa-power-mcp-server:latest"]
    }
  }
}

For Streamable HTTP from a local checkout:

MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 bun run start:http
# http://127.0.0.1:3010/mcp

Prerequisites

• Bun v1.4.0 or higher, or Node.js v24+.

• npm and Docker install native DuckDB dependencies normally on supported platforms. Set CANVAS_PROVIDER_TYPE=duckdb to enable staging and SQL; the default is none.

• The portable Claude Desktop MCPB bundle fixes CANVAS_PROVIDER_TYPE=none and supports inline results. Narrow an oversized request or use npm/Docker for analytics.

Installation

git clone https://github.com/cyanheads/nasa-power-mcp-server.git
cd nasa-power-mcp-server
bun install
cp .env.example .env
bun run rebuild

Configuration

DuckDB startup rejects changes to the four fixed canvas settings. See .env.example for optional transport, authentication, storage, logging and telemetry settings.

Running the server

Local development

bun run rebuild
bun run start:stdio   # or start:http
bun run devcheck
bun run test

Docker

docker build -t nasa-power-mcp-server .
docker run --rm -p 127.0.0.1:3010:3010 -e CANVAS_PROVIDER_TYPE=duckdb nasa-power-mcp-server

The image defaults to HTTP on port 3010. For a client outside loopback, configure authentication and TLS at the deployment boundary.

Project structure

Development guide

See CLAUDE.md and AGENTS.md for architectural rules.

• Handlers throw, framework catches; use ctx.log and ctx.state.

• Register tools in the createApp() array in src/index.ts.

• Validate NASA responses before normalization; preserve sparse fields, source units and annual-period distinctions.

• content[] and structuredContent carry the same complete data.

Contributing

Issues are welcome. Run checks and tests before submitting:

bun run devcheck
bun run test

License

Apache 2.0. See LICENSE.