Microsoft Planetary Computer MCP Tools

A Model Context Protocol (MCP) server that enables GitHub Copilot to interact with Microsoft Planetary Computer Pro.

Overview
Installation
- VS Code
User Guide
- Getting Started
- List of Tools
Reference

Overview

Microsoft Planetary Computer MCP Tools is an MCP Server-compatible service that exposes a set of tools for working with geospatial data on Microsoft Planetary Computer Pro. It provides capabilities for STAC search and discovery, data ingestion, format conversion, error-code troubleshooting, and interactive map-based data selection. Delivered as a VS Code extension, the MCP Server allows users to interact with Planetary Computer Pro entirely through natural language queries via GitHub Copilot, eliminating the need to navigate traditional UI workflows and significantly accelerating geospatial analysis and MPC Pro workflows.

Key Capabilities:

STAC Search & Discovery: Browse and search public datasets from Microsoft Planetary Computer and your private GeoCatalog collections by location, time, and metadata
Collection Management: Create, configure, and clone STAC collections—including render options, mosaic definitions, and thumbnails
PC → GeoCatalog Ingestion: Bulk ingest items from Planetary Computer or external STAC APIs into your GeoCatalog
Natural Language Interface: Interact with Planetary Computer Pro entirely through conversational prompts via GitHub Copilot

Installation

VS Code

Compatible with both the Stable and Insiders builds of VS Code.

Install the GitHub Copilot Chat extension.
Install the Microsoft Planetary Computer Pro MCP Tools extension.

Start (or Auto-Start) the MCP Server

VS Code (version 1.105 or above): You can configure MCP servers to start automatically using the chat.mcp.autostart setting.

Enable Autostart
1. Open Settings in VS Code.
2. Search for chat.mcp.autostart.
3. Select newAndOutdated to automatically start MCP servers without manual refresh.
Manual Start (if autostart is off)
1. Open Command Palette (Ctrl+Shift+P / Cmd+Shift+P).
2. Run MCP: List Servers.
3. Select Microsoft Planetary Computer Pro MCP Tools, then click Start Server.
4. Check That It's Running
  - Go to the Output tab in VS Code.
  - Look for log messages confirming the server started successfully.
(Optional) Configure server behavior in VS Code settings (search for "Microsoft Planetary Computer Pro MCP")

Installation is complete! Microsoft Planetary Computer MCP Tools is now ready to use in VS Code.

User Guide

Prerequisite

The Azure CLI client and an existing Microsoft Planetary Computer Pro GeoCatalog instance must be configured prior to use. Instructions on how to install the Azure CLI are included in this link.

Getting Started

In a PowerShell terminal run az login to log in and select the Azure subscription where your GeoCatalog was provisioned.
Open GitHub Copilot in VS Code and switch to Agent mode.
Click refresh on the tools list
- You should see the Microsoft Planetary Computer MCP Tools in the list of tools
Try a prompt that uses Planetary Computer context, such as How many collections are in the Geocatalog <url>
- The agent should be able to use the Planetary Computer MCP Tools to complete your query
Reference the Microsoft Planetary Computer MS Learn page for information on MPC Pro

List of Tools

✨ Microsoft Planetary Computer MCP Tools function as an assistant layer that augments Microsoft Planetary Computer Pro workflow, helping users to discover, access, and operate geospatial data and services more efficiently. Below is a list of tools, their descriptions, and example prompts to get started:

Discovery & Browsing

Tool Name	Description	Sample Prompt
`list_mpc_stac_collections`	Browse all available public datasets in Microsoft Planetary Computer	"What satellite datasets are available in Planetary Computer Pro?"
`get_mpc_collection_json`	Get detailed metadata about a specific Planetary Computer dataset	"Tell me about the Sentinel-2 collection in Planetary Computer Pro"
`check_mpc_collection_exists`	Check if a collection exists in the public Planetary Computer catalog	"Does the sentinel-2-l2a collection exist in Planetary Computer?"
`list_personal_stac_collections`	List all collections in your private GeoCatalog instance	"Show me all collections in my GeoCatalog"
`get_personal_collection_details`	Get comprehensive details of a specific collection in your GeoCatalog	"Get details for my sentinel-tutorial collection"
`get_personal_collection_json`	Retrieve the STAC collection JSON from your personal GeoCatalog	"Fetch the STAC collection metadata for my-collection"
`check_personal_collection_exists`	Check if a collection exists in your personal GeoCatalog	"Does the naip-2021 collection exist in my GeoCatalog?"

Searching Data

Tool Name	Description	Sample Prompt
`search_mpc_items`	Search for satellite imagery in Planetary Computer by location and time	"Find Sentinel-2 imagery over Seattle from January 2024"
`search_personal_collection_items`	Search for items within your private GeoCatalog collections	"Search my GeoCatalog for imagery near San Francisco"
`search_stac_items`	Search any STAC API with spatial and temporal filters	"Search the Planetary Computer STAC API for Landsat data"

Collection Management

Tool Name	Description	Sample Prompt
`create_personal_stac_collection`	Create a new STAC collection in your personal GeoCatalog	"Create a basic collection called my-imagery"
`create_personal_collection_from_mpc`	Clone a Planetary Computer collection to your GeoCatalog	"Create a collection based on Sentinel-2 from Planetary Computer"
`create_and_ingest_personal_collection_from_mpc`	Full workflow to create and ingest a collection from Planetary Computer	"Copy the NAIP collection from Planetary Computer to my GeoCatalog and ingest the items"
`delete_personal_collection`	Delete a STAC collection from your personal GeoCatalog	"Delete the test-collection from my GeoCatalog"

Collection Configuration

Tool Name	Description	Sample Prompt
`configure_personal_collection_render_options`	Set up visualization presets (Natural Color, NDVI, etc.)	"Configure render options for my Sentinel-2 collection"
`configure_collection_mosaic_definitions`	Define filtering rules for dynamic mosaics	"Set up mosaic definitions for my imagery collection"
`replace_personal_collection_thumbnail`	Add or update a collection's preview image	"Upload a new thumbnail for my naip-data collection"
`delete_personal_collection_thumbnail`	Remove a collection's thumbnail image	"Remove the thumbnail from my test collection"

Ingestion Sources

Tool Name	Description	Sample Prompt
`create_ingestion_source`	Register a cloud storage container as a data source	"Set up an ingestion source for my Azure blob container"
`list_ingestion_sources`	List all registered ingestion sources	"Show all ingestion sources in my GeoCatalog"
`get_ingestion_source_details`	Get details of a specific ingestion source	"Get details for ingestion source abc123"
`delete_ingestion_source`	Remove an ingestion source	"Delete ingestion source xyz789"

Data Ingestion

Tool Name	Description	Sample Prompt
`ingest_stac_item`	Ingest a single STAC item (metadata) into a collection	"Ingest this Sentinel-2 scene into my collection"
`batch_ingest_stac_items`	Batch ingest multiple STAC items into a collection	"Batch ingest these 5 STAC item files into my collection"
`bulk_ingest_stac_items`	Search external API (e.g., Planetary Computer) and bulk ingest results	"Bulk ingest Landsat data from Planetary Computer into my collection"

Ingestion Monitoring

Tool Name	Description	Sample Prompt
`check_operation_status`	Check status of a single ingestion operation	"Check the status of ingestion operation op-12345"
`check_multiple_operations`	Monitor multiple ingestion operations at once	"Check status of all my pending ingestion jobs"

Asset Downloads

Tool Name	Description	Sample Prompt
`download_asset_from_url`	Download a geospatial file from a signed URL	"Download the red band from this Sentinel-2 scene"
`download_spacenet_chips`	Download SpaceNet 5 training chips	"Download SpaceNet Moscow training chips 990-999"

Item Management

Tool Name	Description	Sample Prompt
`delete_stac_item_in_personal_collection`	Delete a STAC item from a collection	"Delete item scene-001 from my collection"

Tool Dispatcher & Discovery

Tool Name	Description	Sample Prompt
`call_mpc_tools`	Universal dispatcher that can call any registered tool by name	"Call the search_mpc_items tool with collection sentinel-2-l2a"
`list_available_tools`	List all tools registered in the MPC PRO Agent server	"What tools are available in this MCP server?"
`get_server_info`	Get comprehensive information about this MCP server and its capabilities	"What can this MCP server do?"

Debugging & Testing

Tool Name	Description	Sample Prompt
`generate_jsonrpc_initialize`	Generate JSON-RPC initialization message for debugging	"Generate the MCP initialization handshake message"
`generate_jsonrpc_tool_call`	Generate JSON-RPC tool call for debugging	"Generate a JSON-RPC message to call search_mpc_items"

Reference

Documentation

Visit Microsoft Planetary Computer Pro MS Learn for information on the Microsoft Planetary Computer Pro platform.

Security

As with all new technology standards, consider doing a security review to ensure any systems that integrate with MCP servers follow all regulations and standards your system is expected to adhere to.

Data Collection

The software may collect information about you and your use of the software and send it to Microsoft. Microsoft may use this information to provide services and improve our products and services. You may turn off the telemetry by following the instructions here.

Microsoft Planetary Computer Pro MCP Tools

Microsoft